1

Lecture Notes for

Methods in Cell Biology

(TRMD 623)







Instructor:
Mark F. Wiser












2
TABLE OF CONTENTS
PREFACECOURSE DESCRIPTION................................................................................7
CHAPTER 1--MICROSCOPY ..............................................................................................9
LIGHT MICROSCOPY.............................................................................................9
ELECTRON MICROSCOPY...................................................................................12
MICROSCOPY APPENDIX. ALDEHYDE FIXATIVES ......................................16
CHAPTER 2--SPECTROPHOTOMETRY..........................................................................19
SPECTROPHOTOMETRY THEORY....................................................................19
INSTRUMENTATION............................................................................................21
APPLICATIONS OF SPECTROPHOTOMETRY..................................................21
FACTORS AFFECTING ABSORPTION...............................................................22
VARIATIONS IN SPECTROPHOTOMETRY.......................................................22
APPENDIX I. PHOTOMULTIPLIER TUBE..........................................................24
APPENDIX II. CALCULATION OF ENZYME ACTIVITY.................................25
CHAPTER 3--FLUORESCENCE........................................................................................29
ENZYME ASSAYS AND FLUOROMETERS.......................................................29
FLUORESCENCE MICROSCOPY.........................................................................30
FLOW CYTOMETRY.............................................................................................33
CHAPTER 4--RADIOCHEMISTRY ...................................................................................35
ATOMIC STRUCTURE AND RADIOACTIVE DECAY......................................35
PROPERTIES OF RADIOISOTOPES ....................................................................36
MEASUREMENT OF -RADIATION...................................................................37
GAMMA-RAY DETECTION .................................................................................42
SAFETY...................................................................................................................42
APPENDIX 1. INTERNAL STANDARDS.............................................................43
APPENDIX 2. CHANNELS RATIO.......................................................................44
APPENDIX 3. MULTIPLE ISOTOPES..................................................................45
CHAPTER 5--pH AND BUFFERS......................................................................................47
DEFINITIONS .........................................................................................................47
HENDERSON-HASSELBACH...............................................................................47
INSTRUMENTATION............................................................................................48
COMPLICATIONS..................................................................................................48
BUFFERING CAPACITY.......................................................................................49
BUFFER SELECTION ............................................................................................49
BUFFER APPENDIX ..............................................................................................50
3
CHAPTER 6--CENTRIFUGATION....................................................................................53
SEDIMENTATION THEORY.................................................................................53
PREPARATIVE CENTRIFUGATION ...................................................................55
CENTRIFUGATION APPENDIX 1. RCF CALCULATION.................................58
CHAPTER 7--INTRODUCTION TO PROTEINS..............................................................61
PROTEIN STRUCTURE.........................................................................................61
PROTEIN STABILITY............................................................................................63
PROTEIN ASSAYS.................................................................................................64
APPENDIX. AMINO ACIDS: PROPERTIES AND STRUCTURE ......................66
CHAPTER 8--DIFFERENTIAL PRECIPITATION OF PROTEINS..................................69
PROCEDURE FOR (NH
4
)
2
SO
4
PRECIPITATION.................................................69
OTHER PROTEIN PRECIPITATION METHODS................................................70
APPENDIX. AMMONIUM SULFATE NOMOGRAM.........................................71
CHAPTER 9--CHROMATOGRAPHY...............................................................................73
BASIC PRINCIPALS...............................................................................................73
EQUIPMENT...........................................................................................................74
ADSORPTION CHROMATOGRAPHY.................................................................75
ION EXCHANGE CHROMATOGRAPHY............................................................75
HYDROPHOBIC CHROMATOGRAPHY.............................................................79
Polar Solvent.........................................................................................................................80
GEL FILTRATION CHROMATOGRAPHY..........................................................80
AFFINITY CHROMATOGRAPHY........................................................................81
CHAPTER 10--MEMBRANES AND DETERGENTS.......................................................83
TYPES OF MEMBRANE PROTEINS....................................................................84
DETERGENTS.........................................................................................................85
DETERGENT REMOVAL......................................................................................87
APPENDIX 1. CELL DISRUPTION TECHNIQUES.............................................88
CHAPTER 11--ELECTROPHORESIS................................................................................91
GEL ELECTROPHORESIS.....................................................................................91
EQUIPMENT...........................................................................................................93
SDS-PAGE...............................................................................................................94
ISOELECTRIC FOCUSING....................................................................................96
TWO-DIMENSIONAL GEL ELECTROPHORESIS..............................................97
PROTEIN DETECTION FOLLOWING ELECTROPHORESIS............................98
PREPARATIVE ELECTROPHORESIS ...............................................................101
ELECTROPHORESIS APPENDIX 1. SIZE CALCULATION. ...........................104
4
ELECTROPHORESIS APPENDIX 2. GEL STAINING......................................105
CHAPTER 12--PROTEIN PURIFICATION OVERVIEW...............................................107
MICROSEQUENCING AND PEPTIDE MAPPING............................................110
APPENDIX. FOLD PURIFICATION AND RECOVERY...................................113
CHAPTER 13--IMMUNIZATION....................................................................................115
IMMUNOGENICITY............................................................................................115
IMMUNOGEN PREPARATION...........................................................................116
IMMUNIZATION..................................................................................................117
COLLECTING AND PROCESSING BLOOD .....................................................120
ANTIBODY PURIFICATION...............................................................................120
CHAPTER 14--MONOCLONAL ANTIBODIES.............................................................123
IMMUNIZATION..................................................................................................123
FUSION..................................................................................................................124
SELECTION...........................................................................................................124
SCREENING..........................................................................................................125
CLONING..............................................................................................................125
PRODUCTION.......................................................................................................126
APPENDIX. B-CELL DIFFERENTIATION........................................................127
CHAPTER 15--IMMUNOASSAYS...................................................................................128
IMMUNOPRECIPITATION .................................................................................129
IMMUNOBLOTTING...........................................................................................131
IMMUNOFLUORESCENCE ................................................................................133
IMMUNOGOLD ELECTRON MICROSCOPY...................................................134
ELISA (AND RIA).................................................................................................135
DIRECT VS. INDIRECT.......................................................................................136
OTHER METHODS...............................................................................................138
CHAPTER 16--NUCLEIC ACID STRUCTURE AND ISOLATION...............................141
ISOLATION OF NUCLEIC ACIDS......................................................................142
ISOLATION OF RNA ...........................................................................................145
DENSITY GRADIENT CENTRIFUGATION......................................................146
ANALYSIS AND QUANTIFICATION................................................................146
APPENDIX 1. COMMON CONVERSIONS ........................................................147
CHAPTER 17--MODIFICATION OF NUCLEIC ACIDS ................................................149
RESTRICTION ENDONUCLEASES ...................................................................149
5
CHAPTER 18--ELECTROPHORESIS OF NUCLEIC ACIDS.........................................153
PULSE FIELD GEL ELECTROPHORESIS .........................................................155
CHAPTER 19--HYBRIDIZATION AND BLOTTING TECHNIQUES............................157
GENERAL PROCEDURES...................................................................................157
FACTORS AFFECTING HYBRIDIZATION.......................................................159
PREPARATION OF LABELED DNA PROBES..................................................160
RFLP AND RESTRICTION MAPPING...............................................................163
IN SITU HYBRIDIZATION..................................................................................164
DNA MICROARRAYS .........................................................................................165
CHAPTER 20--POLYMERASE CHAIN REACTION......................................................167
PCR MECHANISM...............................................................................................167
PRACTICAL ASPECTS........................................................................................168
RNA-PCR...............................................................................................................170
QUANTITATIVE PCR..........................................................................................170
LIMITATIONS.......................................................................................................172
PRECAUTIONS.....................................................................................................174
CHAPTER 21--RECOMBINANT DNA............................................................................175
PREPARATION OF FOREIGN DNA...................................................................175
PLASMID VECTORS............................................................................................178
BACTERIOPHAGE ............................................................................................184
FILAMENTOUS BACTERIOPHAGE..................................................................188
COSMIDS AND YACS .........................................................................................190
QUALITY CONTROL...........................................................................................191
CHAPTER 22--EXPRESSION OF RECOMBINANT PROTEINS...................................193
EXPRESSION IN E. COLI. ...................................................................................193
PREPARATION OF EXPRESSION VECTORS. .................................................195
SCREENING LIBRARIES WITH ANTIBODIES.............................................195
EXPRESSION IN EUKARYOTES .......................................................................196
APPENDIX 1. CODES AND CODONS ...............................................................198
APPENDIX 2. FUSION PROTEINS.....................................................................199
APPENDIX 3. GENERATING FRAMESHIFTS..................................................200
CHAPTER 23--DNA SEQUENCING................................................................................201
DIDEOXY CHAIN TERMINATION....................................................................201
EXTENDED SEQUENCING STRATEGIES........................................................204
MAXIM AND GILBERT.......................................................................................207
CHAPTER 24--SEQUENCE ANALYSIS AND BIOINFORMATICS ............................208
6
SEQUENCE ALIGNMENT...................................................................................208
SEARCHING DATABASES.................................................................................210
PROTEOMICS.......................................................................................................211
APPENDIX 1. WEBSITES.....................................................................................216
APPENDIX 2. RESULTS OF BLAST SEARCH...................................................217

7
PREFACECOURSE DESCRIPTION

This course provides students with a broad overview to the basic biochemical,
molecular and immunological techniques that are commonly used in modern biomedical
research. Lectures will describe the theories and principals behind each of the methods in
addition to discussing the practical aspects and limitations in executing the various procedures.
One of the course objectives is to assist students with their own research by providing them
with sufficient background information so that they are able to design experiments and know
which methods are best suited to address a particular research question or problem. A second
course objective is to provide students a better access to the scientific literature in that a better
understanding of the methods will allow the students to critically evaluate the results and
conclusions of scientific papers. Students anticipating careers involving biological or medical
research at any level will benefit from this course.

The course consists of three sections: 1) general biochemical and biophysical methods,
2) analysis and isolation of proteins and immunological procedures, and 3) analysis of nucleic
acids and recombinant DNA. The first section will cover some basic biochemical procedures
and equipment. Understanding these basic biochemical principals will assist in the
subsequent discussions on proteins and nucleic acids. The section on characterization of
proteins will describe some basic methods used to analyze proteins and provide an overview
on protein purification. In addition, the generation of antibodies and their uses in various
assays will also be covered. The final section on nucleic acids will describe the basic
procedures used in molecular biology including gene cloning, PCR and sequence analysis.

These lecture notes approximately follow the course and are divided into four sections:
1) General Biochemical and Biophysical Methods (Chapters 1-6), 2) Analysis and
Characterization of Proteins (Chapters 7-12), 3) Immunological Methods (Chapters 13-15), and
4) Nucleic Acids and Recombinant DNA (Chapters 16-23). Many of the chapters correspond to
the lectures.
8








PART I



General Biochemical and
Biophysical Methods



Topics covered:

Microscopy
Spectrophotometry
Fluorescence and Flow Cytometry
Radioactivity
pH and Buffers
Centrifugation
9
CHAPTER 1--MICROSCOPY

Cells are small and in almost all situations a microscope is needed to observe them and
their subcellular components. In fact the invention of the microscope led to the discovery and
description of cells by Hooke in 1655. The microscope is still an extremely important tool in
biological research. The light microscope has a limited capability in regards to the size of a
particle that can be examined. The electron microscope provides additional resolution that
allows for the examination of subcellular structures and even molecules.

LIGHT MICROSCOPY

The principal of light microsco-
py is to shine light through a specimen
and examine it under magnification.
The major optical parts of a microscope
are the objective lens, the eyepiece, the
condenser and the light source. The
objective lens functions to magnify the
object. The high degree of magnifica-
tion of the objective lens results in a
small focal length and the magnified
image actually appears directly behind
the objective. The eyepiece functions to
deliver this image to the eye or camera.
Eyepieces also magnify the image, but it
is an empty magnification. In other
words, the eyepiece enlarges the image
but does not increase the ability to see fine details (i.e., the resolution). The condenser functions
to focus the light source on the specimen. The condensor also eliminates stray light and
provides an uniform illumination. An iris diaphragm which controls the amount of light
reaching the specimen is also associated with the condenser lens. In addition, the light intensity
can also be controlled by adjusting the voltage applied to the lamp on some microscopes.

Before using a microscope (Box) it is also important to check
that all of the optical components are centered on an optical axis
so that the best image and resolution are obtained. Aligning the
optical components is usually simple (see instructions manual
for the particular microscope) and needs to be done periodically.
The specimen is then placed on the stage and the objective lens
is focused. The quality of the image produced is highly
dependent on the illumination. The position of the condenser lens is adjusted so that the light is
focused on the specimen and the intensity of the illumination is adjusted. On better microscopes
the illumination can be controlled by both adjusting the diameter of the iris and by adjusting the
voltage applied to the lamp. The amount of illumination is important for controlling resolution
vs. the contrast and the depth of field (see Box for definitions). Resolution and contrast are
antagonistic in that improving one results in a loss of the other. Resolution is increased by

Major components of a light microscope
1. Center light source
and all components
on optic axis.
2. Focus objective.
3. Focus condenser.
4. Adjust illumination.
10
increasing the amount of light. However, the brighter light leads to a loss in contrast. The user
must decide upon the optimal mix of contrast and resolution by adjusting both the voltage (i.e.,
intensity or brightness) of the lamp and the iris diaphragm. The iris diaphragm also has some
effect on the depth of field.

Resolution The ability to discern fine details. Typically expressed as a
linear dimension describing the smallest distance needed
between 2 objects so that both are seen.
Contrast Contrast refers to the number of shades in a specimen.
More shades decreases the contrast, but increases the
amount of information (also called dynamic range).
Depth of Field Refers to the thickness of the specimen that will be in
acceptable focus.

Sample Preparation

Specimens can be examined by simply placing them on a glass microscope slide under a
glass cover slip. However, it is usually necessary to prepare and stain the samples before
examination by microscopy. Fixation is a process by which cells are preserved and stabilized.
Common fixatives include: acids, organic solvents, formaldehyde and glutaraldehyde (see
Appendix for more discussion about aldehyde fixatives). These treatments affix macromole-
cules in position. For example, glutaraldehyde chemically cross-links the primary amines of
neighboring proteins and organic solvents precipitate proteins and other macromolecules.

Thick samples, such as tissues, will need to
cut into thin sections. Following fixation the sample
or cells are embedded into a supporting medium.
Paraffin is a common embedding medium for light
microscopy as well as various plastic resins.
Sectioning is carried out with a microtome (Figure).
The microtome cuts the specimen into thin slices, or
sections, of a specified thickness. It is also possible to
collect the successive slices, called serial sections, and
therefore ascertain the three dimensional aspects of
the tissue or specimen being examined.

The image generated by microscopy depends
upon different components in the sample interacting with and impeding the light waves
differentially. Biological samples are fairly homogeneous (i.e., carbon-based polymers) and do
not greatly impede light. Therefore, it is often necessary to stain cells with dyes to provide more
contrast. Different dyes have different affinities for different subcellular components. For
example, many dyes specifically interact with nucleic acids (i.e., DNA and RNA) and will
differentially stain the cytoplasm and nucleus. These stained subcellular components will
differentially absorb the light waves and result in less light reaching the eyes or camera, and
thus appears darker. Furthermore, since the dyes only absorb certain wavelengths of light, the
11
various structures within the specimen will exhibit
different colors. (See chapter on Spectrophotometry for a
more extensive discussion of chromaphores and light
absorption.)

Variations to bright field (transmission) microscopy

Many modifications of light microscopy that have
specialized applications have been developed (Box). In
dark-field microscopy the specimen is illuminated from the side and only scattered light enters
the objective lens which results in bright objects against dark background. This is accomplished
through the use of an annular aperture that will produce a hollow cone of light that does not
enter the objective lens (see Figure). Some of the light hitting objects within the specimen will
be diffracted into the objective lens (see Figure Inset). The images produced by dark-field
microscopy are low resolution and details cannot be seen. Dark-field microscopy is especially
useful for visualization of small particles such as bacteria.


Dark Field Microscopy Optics Phase Shift vs Diffraction

Both phase contrast microscopy and differential-interference-contrast allow objects
that differ slightly in refractive index or thickness to be distinguished within unstained or living
cells. Differences in the thickness or refractive index within the specimen result in a differential
retardation of light which shifts the phase or deviates the direction of the light (Figure). During
phase contrast microscopy the phase differences are converted to intensity differences by
special objectives and condensers. Normarski optics use special condensers and objectives to
recombine incident and diffracted light waves from a single source at the plane of the image. In
both methods the interference effects between the incident and diffracted light enhance small
differences in the refractive index or thickness of the specimen and leads to an increased
resolution without staining.

In fluorescence microscopy a fluorochrome is excited with ultraviolet light and the
resulting visible fluorescence is viewed. This produces a bright image in a dark background.
Dark Field
Phase Contrast
Differential Interference
Contrast (or Normarski)
Confocal Scanning
Fluorescence
Image Enhancement
12
Confocal microscopy uses the objective lens as both the objective and the condenser. This
allows the illuminating light to be focused on a relatively thin plane. In addition, a pin-hole is
used to further minimize the light coming from other planes. Minimizing the interference from
other planes increases apparent resolution. (Fluorescence and confocal microscopy are
discussed in greater detail in the chapter on Fluorescence.)

Video cameras and image processing have had a major impact on microscopy. Images
are digitized and can be manipulated electronically. This can correct imperfections in optical
systems and can overcome limitations of human eye. In particular, the human eye is not very
effective in dim light and cannot distinguish small differences in intensity against a bright
background. Image enhancement can remedy both of these limitations. However, image
enhancement cannot increase the resolution. This is due in part to the limit of resolution which
is determined by the wavelength () of visible light (see Box).

The theoretical limit of resolution is defined as 0.61/N.A., where N.A. (numerical aperture) is
a property of the objective lens determined by its magnification, diameter and refractive index.
Typical ranges for the N.A. are 0.25-1.32. Visible light has an average wavelength of
approximately 0.5 m making the maximum limit of resolution approximately 0.2 m.
Mitochondria are about the smallest subcellular structures that theoretically can be seen. No
amount of refinement of the optical systems can overcome this physical barrier, even though
the image can be enlarged indefinitely. In addition, the practical limit of resolution will be less
than the theoretical limit of resolution due to optical aberrations in the lenses (see Table).


ELECTRON MICROSCOPY

The relationship between the limit of
resolution and the wavelength of the illumina-
tion holds true for any form of radiation. Thus
resolution can be increased in theory by using
radiation of lower wavelengths. However, the
human eye is only capable of detecting radiation
with wavelengths in the range of 0.4-0.8 m, or the visible spectrum. These problems
associated with the limit of resolution have been overcome by using electrons to generate an
image of the specimen. Particles, such as electrons, travelling near the speed of light behave as a
wave (i.e., radiation) and their effective wavelength is inversely proportional to electron's
velocity. Therefore increased resolution can be achieved by examining a specimen with high
velocity electrons.

The general principal of electron microscopy is analogous to light microscopy (Figure)
except that electrons are used to analyze the specimen instead of visible light. The illumination
source is a white-hot tungsten filament, which emits high velocity electrons. The electron beam
is focused by a condenser lens onto the specimen. The condenser lens, however, is an
electromagnet instead of a glass. These electrons are differentially impeded by the various struc-
tures within the specimen. In other words, some of the electrons are scattered or absorbed by the
atoms of the specimen. The electrons which pass through the specimen are focused with a series
Typical Limits of Resolution for
Common Objective Lenses
Objective
Magnificatio
n N.A.
Theoretica
l (m)
Practical
(m)
4X 0.10 3.05 3.4
10X 0.25 1.22 1.3
40X 0.65 0.47 0.52
100X 1.30 0.24 0.26
13
of magnetic objective lens onto either a photographic plate or a fluorescent screen. The
electrons interact with the photographic plate or fluorescent screen as if they were photons (i.e.,
light) and generate an image. The differential loss of electrons due to the substractive action of
the sample will generate an image in much the same way as the absorption of light creates an
image in light microscopy.

Comparison of Microscope Optics


Sample preparation

It is not possible to view living material with an electron
microscope. Biological samples are usually fixed with glutaralde-
hyde, which cross-links proteins (see Appendix), and treated with
osmium tetroxide, which stabilizes lipid bilayers and proteins.
Osmium tetroxide is reduced by many organic compounds, especial-
ly lipids, which results in cross-linking. Since electrons have very little penetrating power,
samples must be embedded in special plastic resins and cut into thin sections of 0.05-0.1
m. Removing all water from the specimen is necessary for the proper polymerization of the
plastic resin. Following fixation the samples are dehydrated by exposing them to series of
increasing alcohol concentrations until reaching 100%. The dehydrated sample is then put
into a solution containing monomers of the embedding resin and polymerization is induced.
This 'block' containing the sample is sectioned with the ultramicrotome and the ultrathin
sections are place onto copper or nickel grids coated with a thin carbon or plastic film for
support.

Contrast in electron microscopy is dependent upon atomic number of the atoms in
1. Fixation
2. Dehydration
3. Embedding
4. Sectioning
5. Staining
14
the sample. Biological materials, primarily made of carbon, exhibit low atomic number and
exhibit a similar electron scattering as the carbon films on the support grid. To obtain more
contrast, samples are stained with salts of heavy metals, such as osmium, uranium and lead.
Staining can be carried out before the dehydration and embedding or after sectioning.
Different cellular compartments and structures stain differently with the heavy metals.

Electron microscopes are expensive instruments and require substantial training to
operate. Electron microscopy will usually require collaboration with someone having expertise
in electron microscopy. Many universities have shared instrument facilities in which users pay a
fee that includes use of the instrument and technical assistance. Typically fixed samples are
provided to the electron microscopy service for further processing and analysis.

Variations in Electron Microscopy

The standard form of electron microscopy involves
shooting an electron beam through the sample. This is called
transmission electron microscopy, often abbreviated TEM.
Scanning electron microscopy (SEM) detects the electrons
that are scattered by the specimen to form a 3-dimensional
image. The sample is fixed and coated with a thin layer of a
heavy metal such as platinum to form a replica of the specimen. This replica is then scanned
with a thin beam of electrons and the quanitity of electrons scattered along each successive
point of the specimen is measured by detectors which surround the sample (see Figure). Since
the amount of electron scattering depends on the angle and depth of the surface relative to the
beam, the image has highlights and shadows that give it a three dimensional appearance. The
resolution of SEM is not very high (approximately 10 nm with an effective magnification of up
to 20,000 times) and only suface features can be examined. Therefore, the technique is
generally used to study whole cells or tissues.

Transmission (TEM)
Scanning (SEM)
Shadow-casting
Freeze-fracture
Freeze-etching
Negative staining
cryoEM
15
A 3-dimensional appearance
with higher resolutions than SEM can
be obtained by TEM by shadowing.
In this case the metal coating is
applied at an angle resulting in a
replica reflects the height and depth of
the specimen. Shadowing is often
used in conjunction with other
techniques. For example, in freeze-
fracture and freeze-etching cells are
frozen in cryoprotectant and cut with
a knife. Freeze-fracture will often
split the lipid bilayer membranes
which are then shadowed with
platinum. Alternatively, in freeze-
etching, the water is sublimated and
replicas formed.

Negative staining can be
used to visualize macromolecules and
supramolecular structures such as
virus particles or cytoskeletal filaments. The samples are placed on the electron transparent
carbon grids and stained with heavy metals. Areas with biological structures appear more
electron transparent.

The fixation and manipulation of the specimen will often distort cells. Cryo-electron
microscopy is used to overcome this problem. Special holders, which keep hydrated
specimen at -160oC, allows viewing without fixation, staining or dehydration.


Scanning Electron Microscope
16
MICROSCOPY APPENDIX. ALDEHYDE FIXATIVES

Formaldehyde and glutaraldehyde are a commonly used fixatives for both light and
electron microscopy. Formaldehyde is a small molecule (HCHO). The formaldehyde
monomers form polymers in aqueous solutions. The liquid known as formalin is 37-40%
formaldehyde by weight and most of the polymers are 2-8 units long. Higher polymers (n up
to 100) are insoluble in water and sold as a white powder called paraformaldehyde. To be
useful as a fixative, the solution must contain monomeric formaldehyde. Dilution of
formalin with a buffer at physiological pH results in an almost instantaneous formation of
monomers. Conversion of paraformaldehyde to monomers requires heat (typically 60
o
C) and
the addition of hydroxide ions. Commercial formalin contains about 10% methanol and
small amounts of formate ions, whereas a formaldehyde solution prepared from paraform-
aldehye initially does contain any methanol or formate.

Formaldehyde's mechanism of action is based on the reaction of the aldehyde group
with primary amines in proteins. A cross-link between neighboring proteins can also be
formed if the primary amines are close enough together. The initial reaction of
formaldehyde with protein is complete within 24 hours, but the formation of cross-links,
called methylene bridges, proceeds much more slowly (several weeks). Lipids, nucleic acids
and carbohydrates are trapped in a matrix of insoluble and cross-linked proteins. In practical
terms, formaldehyde penetrates tissues rapidly (because of its small size), but it slowly
cross-links the proteins.



Glutaraldehyde contains two aldehyde groups separated by three methylene bridges
(HCO-[CH
2
]
3
-CHO). These two aldehyde groups and the flexible methylene bridge greatly
increases the cross-linking potential of glutaraldehyde over formaldehyde. In solution
glutaraldehyde exists as polymers of various sizes which exhibit an enormous potential for
cross-linking proteins (Figure). In contrast with formaldehyde, the chemical reaction of
glutaraldehyde with protein is fast, but the penetration of tissue is slower, especially for the
17
larger oligomers. Therefore, an 'EM grade' glutaraldehyde, which contains low polymers,
should be used for fixation. In addition, fixation with glutaraldehyde results many leftover
free aldehyde groups which cannot be washed out of the tissue. For many applications these
free aldehyde groups needs need to be removed or blocked. A common blocking method is
to treat with glycine or another small primary amine.

The combination of formaldehyde and glutaraldehyde is also used as a fixative for
electron microscopy. This takes advantage of the rapid penetration of formaldehyde
molecules, which quickly stabilizes the structure of the tissue, followed by a more thorough
cross-linking of proteins mediated by the more slowly penetrating glutaraldehyde.



19
CHAPTER 2--SPECTROPHOTOMETRY

Spectrophotometry is a versatile analytical tool. The underlying principle of spectro-
photometry is to shine light on a sample and to analyze how the sample affects the light.
Advantages of spectrophotometry are: 1) it is often non-destructive (i.e., can measure and
recover sample), 2) it is selective (often a particular compound in a mixture can be measured
without separation techniques), 3) it has a short time interval of measurement (10
-14
seconds).

SPECTROPHOTOMETRY THEORY

Light can be described as a wave. This wave has an electric component and a magnetic
component which are perpendicular to each other (Figure). Electromagnetic radiation exhibits a
direction of propagation and wave-like pro-
perties (i.e., oscillations). The energy of
electromagnetic radiation is defined as:

E = hc/ = h

where E = energy, h = Planck's constant, c =
the speed of light, = the wave length, and =
frequency. Light behaves both as a wave and as a particle. The conceptual particle of light is
called a photon and is represented by h. Electromagnetic radiation exhibits a wide spectrum
and specific ranges of wavelengths have names (Figure). The energy of electromagnetic
radiation is inversely proportional to its wavelength.

When a light wave encounters a particle, or molecule, it can
be scattered (i.e., direction changed), absorbed (energy transferred),
or unaffected. Molecules only absorb discreet packets of energy, or
quanta, and absorption occurs when the energy of the photon corre-
sponds to differences between energy levels in that particular mole-
cule. These discrete energy levels, called electronic energy levels,
are a property of the particular molecule and are determined by the
spatial distribution of the electrons.

Absorption of the energy from the photon elevates the molecule to an excited electronic
state (see Figure in Chapter 3) by causing an electron to move from one orbit to another. These
electronic energy levels are further subdivided into vibrational levels. The vibrational levels
correspond to stretching and bending of various covalent bonds. The transitions to the excited
20
state can occur between different vibrational levels giving a range of energy that can be
absorbed by the molecule.

A molecule or substance that absorbs light is called a chromophore. Chromophores
exhibit unique absorption spectra (Figure) and can be defined by a wavelength of maximum
absorption, or
max
, of a broad absorbtion band due to the vibrational levels. The absorption
spectra can consists of several absorption maxima of various amplitudes. A large number of
biological molecules absorb light in the visible and ultraviolet (UV) range.



The net affect of absorption is that the intensity of the light decreases as it passes
through a solution containing a chromophore. The amount of light absorbed depends on the
nature of the chromophore, the concentration of the chromophore, the thickness of the
sample, and the conditions (eg., pH, solvent, etc.) under which absorption is measured.
Absorption is governed by the Beer-Lambert Law:

I = I
o
10
-dc
or log(I/I
o
) = -dc

where I = final light intensity, I
o
= initial light intensity, = molar extinction coefficient, d =
thickness, and c = molar concentration. Absorption (A) will be defined by:

A = -log(I/I
o
) = dc

The molar extinction coefficient () is defined as the A of 1 M of pure compound under
standard conditions and reflects something about the nature of the chromaphore. The units of
are liter/cmmole. However, the extinction coefficient can be expressed in other units. For
example, it can be expressed in terms of mM concentration. The thickness of the sample (d) is
almost always 1 cm and therefore can be ignored in calculations. Sometimes, though, the
extinction coefficient units is expressed in cm
2
/mole (by converting liters to cubic centimeters)
and care should be taken in making calculations. In cases where the molecular weight of the
substance is not known, or varies, E
1%
is used as the extinction coefficient. E
1%
is defined as the
A of a 1% (w/v) solution. It is important to precisely record the units of when looking it up or
determining it experimentally since these units will determine the concentration. It is also
important to record the conditions (eg., pH, solvent, temperature, etc.) for an extinction
coefficient (see below).
21

INSTRUMENTATION

Spectrophotometers produce
monochromatic light and then accurate-
ly measure the light intensity. The major
components of a spectrophotometer are
the light source, a monochromator,
sample holder, a light detector (photo-
tube), and a meter. In most instruments a
tungsten lamp is used for the visible range and either high pressure H
2
or D
2
lamps are used for
UV range. Monochromatic light is generated by either 1) a movable prism, 2) a diffraction
gradient, or 3) filters. Monochromatic light is projected through the sample and then measured
by a photomultiplier tube. A photomultiplier tube converts the energy of the light photons into
electrons (see appendix). The voltage resulting from these electrons is measured by a meter and
converted to an absorbance value. The I
o
(initial intensity) is determined by calibrating the
instrument with a 'buffer blank'. The relative difference in the light intensity between the blank
and the sample is then expressed as the absorbance (A). Spectrophotometers often include
accessories such as chart recorders or microprocessors for data analysis.

APPLICATIONS OF SPECTROPHOTOMETRY

Determining the concentration of substances in solution is the most common use of the
spectrophotometer. Exact concentrations can be determined in cases where are known and the
measurement is carried out under the prescribed conditions. The substance being measured does
not necessarily need to absorb radiation if it can scatter radiation. For example, measuring the
A
600
is a quick and easy way to monitor bacterial growth and determine the number of bacteria
in cultures. In addition, since compounds exhibit unique absorption profiles, spectrophotometry
can also be used to identify unknown compounds.

Spectrophotometry is also a convenient method to measure enzyme activity in cases
where the substrate and the product exhibit different
max
. Either the disappearance of substrate
or the appearance of product over time is measured. The change in the A

per unit time

(generally per minute) is calculated. The change in A

of a blank (= identical sample without

enzyme source) is subtracted from this value. The enzyme activity in terms of amount of
product formed per unit time per mg protein can be calculated by factoring in the amount of
enzyme, dilution factors and the extinction coefficient (see example in appendix). A typical
example of a formula for the calculation of enzyme activity is:

activity = (A
sample
- A
blank
)volume10
6
/timemg protein

where activity is expressed as moles product formed/min/mg protein; A
sample
is the change in
absorbance of the sample containing enzyme; A
blank
is the change in absorbance of a sample
containing everything except the enzyme; the volume in the cuvette expressed in the same units
as ; 10
6
moles per mole (assuming is expressed in moles); is the molar extinction
coefficient; time is the time in minutes the reaction was measured; and mg protein in the
Basic Spectrophotometer Components
22
cuvette. In deriving such formulas it is important to match the units. The units of may also
include the 1 cm thickness of the cuvette which is ignored in the calculations.

FACTORS AFFECTING ABSORPTION

Although the absorption spectrum is primarily
determined by the chemical structure of the chromophore,
environmental factors can affect
max
and . The pH determines
the ionization of chromophore which in many cases will affect
the absorption properties of a chromophore. Indicator dyes and
pH paper are examples of this phenomenon.

The polarity of the solvent or neighboring molecules
can also affect absorption. Because of this effect, spectro-
photometry can also be used to determine structural features of
macromolecules. For example, whether particular amino acid
residues are buried within the internal portion of proteins or
exposed on the aqueous solvent at the protein surface can be
determined by spectrophotometry.

The relative orientation of neighboring chromophores
also affects absorption. Hypochromism of nucleic acids
(Figure) is an example of orientation effects. The absorbance (i.e., ) decreases as free
nucleotides (dotted line) are polymerized into single-stranded DNA (dashed line) and decreases
further in double-stranded DNA (solid line).


VARIATIONS IN SPECTROPHOTOMETRY

It is also possible to use other spectra besides UV/visible range. For example, vibrations
between the atoms of molecules can be analyzed using infrared (IR) and Raman spectro-
photometries. Many molecules will exhibit characteristic 'signatures' and therefore can be
identified.

Light can be polarized so
that all of the waves are in the same
orientation. The study of the absorp-
tion of polarized light can yield more
information about the structure of
molecules if the chromophores have
optically active centers. Circular
dichroism (CD) measures the ability
of chromophores to differentially absorb left and right circularly polarized light. Optical rotary
dispersion (ORD) measures the ability of a optically active chromophore to rotate plane
polarized light. Both CD and ORD are useful in structural studies or proteins and nucleic acids.
For example, it is possible to approximate the amount of -helix, -sheet and random coil in

Spectra of T7 DNA
Other Forms of Spectrophotometry
Spectrophotometry Comment
Infrared (IR)
Raman
vibrational
levels
Optical Rotary Dispersion (ORD)
Circular Dichroism (CD)
polarize
light
Nuclear Magnetic Resonance (NMR)
Electron Spin Resonance (ESR)
magnetic
moments
23
proteins.

The effects of molecules on the magnetic component of radiation can also be analyzed.
Nuclear magnetic resonance (NMR) is a spectroscopic method capable of yielding information
on the structure of molecules, interactions between molecules and molecular motion. This
method is based upon the principle that a spinning charge (i.e., the nucleus) generates a
magnetic field. Similarly, an electron also possesses a spin magnetic moment which can be
analyzed by electron spin resonance (ESR). A common use for ESR in biological sciences is to
monitor the fluidity of membranes.
24
APPENDIX I. PHOTOMULTIPLIER TUBE

Photomultiplier tubes, also called photoelectric cells, convert light intensity into an
electrical current. The basic principle of a photomultiplier tube is the ejection of electrons
from a metal surface and the measurement of the resulting current (Figure). A receiver (K) is
coated with potassium or other metal in a highly evacuated tube. The tube also contains a
wire screen (W) which is connected to the receiver through a battery (B) and a sensitive
galvanometer (G). Light hitting the receiver ejects ejects electrons which are attracted to the
positive charge of the screen. The resulting current is directly proportional to the number of
electrons ejected per unit time, which is proportional to the light intensity. In other words,
the energy of the photon (h) is transferred to a single electron within the metal. If this
energy is sufficeintly large then the electron will be ejected from the surface of the metal
and retain some kinetic energy which depends upon the energy of the photon that ejected it.
The number of electrons ejected will depend on the number of photons (i.e., light intensity).
This is known as the photoelectric effect.


25
APPENDIX II. CALCULATION OF ENZYME ACTIVITY

Acetylcholine esterase (ACHE) removes the acetate group from choline. This activity can be
measured with the non-natural substrate acetylthiocholine. ACHE cleaves the sulfide bond
producing a free thiol (thiocholine) and acetate. This free thiol will spontaneously react with
dithionitrobenzoate (DTNB) to produce 5-thionitrobenzoic acid (5TNB), which has a red
color. Therefore ACHE activity can be measured using the following coupled reaction:

acetylthiocholine thiocholine + acetate (catalyzed by ACHE)

thiocholine + DTNB 5TNB + oxidized thiocholine (occurs spontaneously)

You have a sample with a protein
concentration of 0.34 mg/ml and
would like to calculate the activity of
ACHE (units = moles product
formed/ minmg protein). Thirty l of
the sample are added to a cuvette in
final volume of 1 ml containing all
the necessary substrates. The A
405
is
plotted over time as depicted by the
figure on the right. The A
405
increases
from 0.015 to 0.287 in 10 minutes.
The A
405
of a blank (containing all the
same reagents except the protein
sample) goes from 0.014 to 0.073 in 10 minutes. The
405
for 5TNB is 13.3x10
3

liter/cmmole

Calculations and Explanations

The following formula (see derivation below) can be used to calculate the enzyme activity:

(A
sample
- A
blank
)cuvette vol.10
6
/timesample vol.protein conc.

A
sample
This represents the change in the absorbance from the beginning to the end of the
measurement period. Obtained by subtracting 0.015 from 0.287.

A
blank
Some DTNB will spontaneously convert to 5TNB and this needs to be substracted
from sample values. Obtained by subtracting 0.014 from 0.073.

cuvette vol Absorbance is proportional to concentration and must be multiplied by the volume
to calculate the absolute number of moles product formed. The units must match
; liters in this example.

10
6
This converts the moles of to moles. The conversion factor will change
according to the desired final units and units of .
0.287
0.015
0.073
0.014
0.0
0.1
0.2
0.3
0 5 10
minutes
A405
sample
blank

26

The molar extinction coefficient converts absorbance values to concentrations.
represents an inherent property of the chromophore and is usually looked up. It is
important that the conditions of assay are similar to the conditions used to
determine . The thickness unit (cm) can be ignored since the standard cuvette
size is 1 cm.

time Enzyme activity is expressed per unit time (usually minutes) and this represents
the time interval the absorbance was measured. If the times for the samples and
blank do not match then it will be necessary to calculate (A
sample
/time -
A
blank
/time).

sample vol All of the parameters above represent the activity in the cuvette. Therefore it is
necessary to divide by the sample volume so that the activity is expressed per unit
volume of the sample.

protein Dividing by the protein concentration will provide information about the amount
of enzyme activity per unit protein (i.e., specific activity).

Plugging the values:

(0.272 - 0.059)(0.001 L)(10
6
mole/mole)/(13.3x10
3
L/cmmole)(10 min)(0.03 ml)(0.34
mg/ml) =

0.16 moles thiocholine formed per min per mg protein


Example of formula derivation.

There are several strategies and formulas that can be used to calculate enzyme activities. All
of them start with the Beer-Lambert Law:

A = dc

rearranging the equation to solve for concentration and adding the thickness results in:

c = A/(1 cm)

The 1 cm will not affect the calculation and can be ignored as long as the units of are
corrected to account for this. Measuring enzyme activity involves determining the change in
concentration over time. In addition, the spontaneous formation of product (blank without
enzyme) needs to be subtracted from the sample resulting in:

c/time = (A
sample
/time - A
blank
/time)/

If the sample and blank are measured for the same time periods the equation can be convert
27
to:

c/t = (A
s
- A
b
)/t

Multiplying both sides of the equation by the cuvette volume (same units as ) and 10
6

moles/mole (assuming is expressed in moles) will result in:

mole product formed/min = (A
s
- A
b
)(vol.)10
6
/t

This formula will represent the total enzyme activity in the cuvette. Dividing by the volume
of sample used in the assay will give the amount of enzyme activity per unit volume of the
original sample:

mole/minml = (A
s
- A
b
)(vol.)10
6
/t(sample vol.)

Alternatively enzyme activity is often expressed per unit of protein (or specific activity) and
therefore needs to be divided by the amount of total protein added to the cuvette instead of
the sample volume, or:

mole/minmg protein = (A
s
- A
b
)(vol.)10
6
/t(mg prot.)

The mg protein can be calculated by multiplying the protein concentration (in mg/ml) by the
sample volume (in ml).

Additional Practice Problem

Dehydrogenase activities are easily measured in spectrophotometric assays because of
strong absorbance of reduced flavin nucleotides (NADH and NADPH) at 340 nm as
compared to the oxidized nucleotides (NAD
+
, NADP
+
). The
340
of either NADH or
NADPH is 6.22 x10
3
liter/cmmole.

Calculate the amount of glutamate dehydrogenase (GDH) activity expressed in moles
glutamate formed per min per mg protein using the following reaction parameters: the
protein concentration in the sample is 0.23 mg/ml; 0.01 ml of sample was measured in a
final volume of 1 ml; the A
340
decreased from 0.60 to 0.52 in 10 minutes; and the A
340
of a
blank (containing no enzyme) decrease from 0.60 to 0.59 in 10 minutes. GDH catalyzes the
following reaction:

-ketoglutarate + NH
4
+
+ NADPH glutamate + NADP
+


[Answer is 0.49 moles glutamate/minmg protein]

28

29
CHAPTER 3--FLUORESCENCE

Molecules absorbing the energy of electromagnetic radiation (i.e., photons) will be
elevated to a higher energy level, or excited state. These excited molecules will return to the
ground state and some molecules emit radiation on their return to the ground state. This
phenomenon is known as fluorescence and fluorescent molecules are known as fluorochromes.
Fluorochromes have distinct absorption spectra as well as emission spectra. The wavelengths of
the emitted radiation are longer than the absorbed wavelengths (i.e., lower energy).



Fluorescence is widely used in the biological research. In
particular, many molecules absorb in the ultraviolet range (UV)
and emit radiation in the visible range. These fluorochromes can
be used as sensitive probes to investigate biological phenomenon
and are utilized in many different applications (Box).

ENZYME ASSAYS AND FLUOROMETERS

Enzyme substrates are available in which either
the product or the substrate is fluorescent. As in the
case of spectrophotometry, this provides a convenient
method to either follow the appearance of product or
the disappearance of the substrate. In general
fluorescent substrates provide greater sensitivity than
conventional spectrophotometric assays. The amount
of fluorescence is measured using a fluorometer. The
fluorometer is similar to a spectrophotometer, except
that the photomultiplier tube to detect the emitted light is located at a right angle from the path
of the excitation light (Figure). The sample is exposed to UV light of the desired wavelength
and emitted light of the appropriate wavelength range is detected by a photomultiplier tube. The
intensity of the emitted light is quantified and usually expressed as relative fluorescence units.
enzyme assays
detection of nucleic
acids (gels)
microscopy
flow cytometry
30

FLUORESCENCE MICROSCOPY

Since many fluorescent molecules emit light
in the visible range it is possible to view
fluorescence in conjunction with microscopy. An
UV light source is used to illuminate the sample
through the objective lens using a beam-splitting
mirror. The fluorescence emitted from the sample
(epifluorescence) passes through this same mirror,
but the UV light does not. Filters before the beam-
splitting mirror will control the excitation wave-
length and filters before the eyepiece (or camera)
will control the wavelength of the emitted light.

Fluorescent probes can be used to identify individual
cells or subcellular components (Box). The location of fluoro-
chromes associated with cells will appear as light objects
against a dark background. Immunofluorescence (discussed in
greater detail in immunoassays) uses antibodies with a
conjugated fluorescent molecule (eg., fluorescein or rhoda-
mine) to determine the location of a protein within a cell.
Numerous fluorescent dyes which bind to specific subcellular
components are also available. For example, acridine orange
fluoresces when it binds to either RNA and DNA. Acridine
orange bound to RNA fluoresces a yellow to green color and
acridine orange bound to DNA fluoresces orange to red. Other
fluorescent probes which bind to specific subcellular compart-
ments are also available. In addition, there are many fluores-
cent probes available that are sensitive to pH, divalent cations such as Ca
2+
, and membrane
potentials. Such probes permit physiological measurements in individual living cells.

Dual-labeling experiments in which fluorochromes with different emission spectra are
co-incubated with the sample can also be carried out. The different fluorochromes will appear
as different colors due to the differences in their emission spectra. If the emission spectra are far
enough apart then it is possible to use filters so that the two fluorochromes are examined
separately. Dual-labeling experiments are especially useful for determining whether two
proteins (or other substances) are localized to the same subcellular compartment.

Confocal scanning laser microscopy (CSLM). The confocal microscope increases the
resolution, as compared to conventional fluorescent microscopy, by minimizing the light from
out of focus planes. In CSLM the illuminating light, a laser, is focused onto the specimen by the
objective lens. Light reflected back from the illuminated area is collected by the objective and
directed through a pinhole placed in front of the detector. This diameter of the pinhole can be
controlled and determines the thickness of the focal plane. This confocal pinhole rejects light
that did not originate from the focal plane (see Figures). A thinner optical plane (i.e., smaller pin
ethidium bromide
(DNA/RNA)
DAPI (only DNA)
rhodamine 123
(mitochondria)
C
5
-CMB-ceramide
(Golgi)
free Ca
2+
indicators
pH indicators
membrane potential
indicators
antibodies against
specific proteins
31
hole diameter) provides a higher degree of resolution since interference from other planes is
minimized.




32
The image is produced by scanning the optical field pixel by pixel (x, y dimensions)
with the laser and collecting the fluorescence intensity with a photomultiplier tube. A computer
generates the image from this fluorescence data. In the case of dual- or multiple fluorchromes
the data on the different wavelengths is collected separately and assigned different colors.
Images of the separate fluorochromes or a merge image can then be displayed or exported into a
file.

It is also possible to section a specimen into several optical planes by moving the stage
vertically (z-dimension). This requires a computer-controlled motor to precisely move the stage
according to the specified distances. The process of making these 'optical sections' involves
setting the focal plane at the bottom of the specimen, recording the image in digital form, and
moving the stage up a prescribed distance. Another digital picture is taken and the process is
repeat throughout the thickness of the specimen resulting in images of consecutive focal planes
separated by a specific interval. These digitized images are stored and can be viewed in
sequence. In addition, three-dimensional images can be reconstructed from the optical sections
through the use of computers.
33

FLOW CYTOMETRY

Fluorescent microscopy provides qualitative information about individual cells. How-
ever, obtaining quantitative data is difficult and labor intensive. Fluorometry provides quantita-
tive data, but represents the average of the population being analyzed. In other words, fluorom-
etry does not reveal information about heterogeneity within that population. Flow cytometry can
provide quantitative information on individual cells.

The principal of flow cytometry is to pass cells (or particles), in single file through a
laser beam and analyze the interaction of the laser with individual cells (Figure). For example,
the scatter (i.e., direction change) of the laser provides information about the size and shape of
the cell. In addition, cells can be labeled with fluorochromes which are excited by the laser. The
amount of fluorescence associated with the cells provide quantitative data on specific target
molecules or their subcellular constituents.

Flow cytometry can quantify virtually any cell-associated factor or organelle for which
there is a fluorescent probe (or natural fluorescence). For example, the amount of DNA per cell
can be measured and provide information about the proportions of cells in various phases of the
cell cycle (Figure). Through the use of multiple detectors and lasers it is also possible to analyze
two or more fluorochromes with different emission spectra simultaneously. The data for the
various cellular parameters being measured are accumulated for each cell and sub-populations
can be identified and characterized. For example, the data can be plotted as dots defined by the
fluorescence values of the two fluorochromes (Figure). The number of cells with particular
characteristics are tabulated and their distributions in the cell population determined.

34


In addition, many flow cytometers are equipped with a cell sorter, or a fluores-
cence activated cell sorter (FACS). Cell sorting is accomplished by passing the cells between a
charged plate and applying a pulse of current to cause the cell to change directions. In this way
cells with particular characteristics (i.e., predetermined levels of fluorescence, scatter, etc.) can
be separated from a mixture of cells. The separated cells can then be subjected to other analyses.
35
CHAPTER 4--RADIOCHEMISTRY

Radioactivity is widely used in biological research. Radioisotopes are used to label
molecules of interest and these molecules can be analyzed. One advantage of radioactivity, as
compared to other analytical techniques, is its sensitivity. For example, detecting 10
-12
moles
(pmole) is feasible in many applications. The other major advantage relates to the ability to
label molecules with radioisotopes and not greatly affect their chemical properties. The use of
chromophores or fluorochromes for analytical work usually involves adding another chemical
group to the molecule of interest which will have an impact on the properties of the molecule.
These two features, sensitivity and chemical similarity, make radioisotopes powerful experi-
mental tools. For example, its is possible to follow the conversion of one molecule into another
through the use of 'tracer' techniques. Radiolabled compounds can also be used as ligands in
binding studies or as substrates (eg., enzyme assays, macromolecular synthesis or permeability
studies).

ATOMIC STRUCTURE AND RADIOACTIVE DECAY

Atoms are composed of a nucleus, containing protons and neutrons, surrounded by
electrons. Protons (p
+
) have a mass of 1 atomic mass unit (amu) and a charge of +1, neutrons (n)
have a mass of 1 amu and are neutral (i.e., no charge), and electrons (e
-
) have a charge of -1 and
a negligible mass (0.0005 amu). The mass of an atom is equal to the sum of the number of
protons and neutrons. An atomic mass unit is defined as the mass of a proton or neutron and
approximately equals 1.66 x 10
-24
grams. The number of electrons orbiting the nucleus is equal
to the number of protons.

The atomic number of an atom is equal to the
number of protons and defines the chemical element.
Isotopes are species of the same chemical element with
different atomic masses (i.e., a different number of
neutrons). Isotopes can be either stable or unstable.
Unstable isotopes will undergo nuclear rearrangements, or decompose, and radiation is emitted
as part of this decomposition. This emitted radiation can be the ejection of subatomic particles
(i.e., neutrons, protons, electrons) from the nucleus or the emission of electromagnetic radiation
(Table). Alpha-radiation is not widely used in biomedical research since the relatively large -
particles are difficult to detect.

Radioisotopes are found in nature and any particular chemical element is composed of a
mixture of isotopes both radioactive and non-radiactive. Radioisotopes can also be synthesized.
For example, bombarding nitrogen with neutrons (n) will displace a proton (p) and result in the
formation of radioactive carbon (
14
C) according to:

14
N
7
+ n
14
C
6
+ p
+


The superscripts are the atomic masses (neutrons + protons) and the subscripts are the atomic
numbers (i.e., protons). The resulting
14
C is unstable and will decay by converting a neutron
into a proton. (A neutron is composed of a proton plus an electron and this nuclear electron is
Major Types of Radiation
-particles 2p2n (He nucleus)
-particles electron (e
-
)
-rays photon (h)
Only - and -radiation are widely
used in biological research.
36
ejected as a -particle). Conversion of a neutron into a proton results in the formation of
nitrogen according to:

14
C
6

14
N
7
+ (or e
-
) + neutrino

Also emitted from the decaying nucleus is a neutrino, an entity with little mass and no charge.


PROPERTIES OF RADIOISOTOPES

Each radioactive decay, or emission,
has a specific energy. These emissions have
distinct energy spectrums, or ranges for each
particular isotope (Figure). In the case of b-
radiation this energy corresponds to the
velocity at which the electron is ejected from
the nucleus. Each isotope is characterized by a
maximal energy, or E
max
, that can be emitted
and an average energy of the emissions. The
average energy of the emissions is approximately one-third of the E
max
.

Radioactive decay is a random process in
that the probability that a particular atom will
decay is the same for each of the atoms of that
particular isotope (i.e., one decay does not affect
another decay). However, each isotope has a dis-
tinctive rate of decay which is defined by a half-
life. The half-life of an isotope is the time it takes
for half of the radioactivity to decay and is inde-
pendent of the amount of radionuclide. Radioiso-
topes can be defined by the energy of their emissions and their half-life (Table).

Units of radioactivity are expressed as disinte-
grations (or decays) per unit time. For example, the
amount of radioactivity in a sample can be expressed as
disintegrations per second (dps) or disintegrations per
minute (dpm). The recommended unit of radioactivity
is the Becquerel (Bq) which is defined as one dps
(Table). However, the curie (Ci) is still commonly used. The definition of one curie was origi-
nally the dps in one gram of
226
Ra. This definition has been standardized to 3.7 x 10
10
dps.
Radioactivity is often expressed as counts (= detected disintegrations) per unit time (i.e., cps or
cpm). All of these units express how much radioactivity is present. Because each disintegration
represents a loss of radioactivity, this value changes with time according to the half-life of that
isotope. In other words, there is only half of the amount of radioactivity remaining after one
half-life.


Energy spectra of
3
H and
14
C. On the y-axis is
the frequency, or number of emissions with a
particular energy (x-axis).
Properties of Commonly Used
-emitters in Biological Sciences
Isotope *E
avg
*E
max
Half-Life
3
H 0.0055 0.015 12.3 years
14
C 0.05 0.15 5500 years
32
P 0.7 1.71 14.3 days
35
S 0.0492 0.167 87.1 days
*Energy expressed in MeV (mega electron
volts)
Units of Radioactivity
Curie (Ci)
3.7 x 10
10
dps
2.2 x 10
12
dpm
becquerel (Bq) 1 dps
The ICRU recommends the use of the Bq
as the official unit of measure.
37
Two other terms used to describe radioactive compounds are specific activity and
radiochemical purity. The specific activity refers to the disintegration rate per unit mass (eg.,
GBq/mmole) and reflects the mixture of non-radioactive and radioactive isotopes. The
maximum possible specific activity is determined by the half-life. In other words, it is an
inherent property of the isotope. The specific activity provides a means to convert units of
radioactivity into mass units (eg., mg, moles, etc).

Radiochemical purity refers to the variety of compounds that are radiolabeled in the
preparation. This is expressed as a percent and indicates the degree of purity. This can change
over time due to decomposition of the compound of interest into other compounds.


MEASUREMENT OF -RADIATION

Three methods are commonly used to
measure -radiation (Box). None of these
methods measure -particles directly, but measure the effect of
collisions between -particles and some component of the assay
system.

Geiger-Mueller Counter

The G-M counter is not widely used as a quantitative
instrument, but is used most often for monitoring possible
radioactive contamination and estimating the amount of
radioactivity. A typical G-M counter consists of a cylindrical
cathode with a central anode. The tube is filled with an inert gas
and sealed on one end by a mica sheet. -particles passing
through the mica sheet will dislodge electrons from the inert gas
resulting in ion pairs. The ion pairs are then detected as an
electrical potential between the anode and cathode. The G-M counter is unable to detect low
energy isotopes since radiation needs to penetrate the mica sheet and dislodge enough electrons
to create a detectable electrical potential.

Autoradiography

Autoradiography is the exposure of x-ray film (or a photographic emulsion) to a
radioactive sample. The radioactivity behaves as light and results in the exposure of the
photographic emulsion on the film. Radiation activates the silver halide in the photographic
emulsion. During development the activated silver halide is converted to metallic silver. The
metallic silver will appear as 'grains' on the film in positions where radioactivity is located. In
other words the exposed film will be darken in proportion to the amount of radioactivity it was
exposed to. If the film is juxtaposed to the sample it is possible to precisely localize the position
of the radioactivity.

Geiger-Mueller Counting
Autoradiography (film exposure)
Scintillation Spectrophotometry
38
The two most common uses of autoradiography is
in conjunction with gel electrophoresis or in conjunction
with microscopy. Autoradiography of gels allows the
identification and quantitation of specific proteins or
nucleic acids (discussed later). It is also possible to
localize radioactive markers to particular cells in tissue
sections or subcellular structures within cells by auto-
radiography. The typical procedure is to cover cells with a photographic emulsion (Box) and
develop the emulsion after sufficient exposure. Dark grains (or spots) will appear over the cells
or structures which contain the radioactive marker. It is usually desirable to use lower energy
isotopes in autoradiography. High energy isotopes will travel greater distances and activate
more silver halide atoms thus producing long 'tracks'.

Scintillation Spectrophotometry

The scintillation spectrophotometer efficiently detects and measures low energy
isotopes and low levels of radioactivity. In liquid scintillation counting (LSC) the radioactive
sample is mixed with a scintillation cocktail containing chemicals known as fluors. The -
particles collide with solvent molecules and transfer energy to these solvent molecules. The
excited solvent molecules will then transfer this excess energy to the fluors (= fluorochromes).
The excited fluors will emit a flash of light (i.e., fluoresce) when they return to its ground
(unexcited) state. This flash of light is then detected and evaluated with photomultiplier tubes.
The scintillation spectrophotometer records the number of flashes of light, which are converted
to counts per minute (cpm).


Instrumentation. The instrument typically consists of two or more photomultiplier tubes
(Figure). As in the case with spectrophotometers the photomultiplier tubes will convert light
into an electrical signal. In addition, the scintillation counter has coincident circuits which only
count events that are simultaneously recorded on both photomultiplier tubes (signal
summation), thus minimizing background counts from the photomultiplier tubes.

-particles with higher energy levels will excite more solvent molecules which will
activate a greater number of fluor molecules. This will result in more photons being emitted
which generates an electrical pulse with a higher voltage. Therefore, not only can radioactive
disintegrations be counted, but the energy of each disintegration can also be determined.
1. Fix tissue or cells to slide.
2. Cover with a photographic
emulsion.
3. Expose and develop.
4. Examine under microscope.
39
Through the use of channel discriminators this allows the number of counts within particular
energy ranges to be determined.



Background. Background radiation sources (Box) can interfere
with the accurate measurement of radioactivity in experiments. The earth
is constantly being bombarded by cosmic radiation and radioactive
isotopes are found in nature. Cosmic and other natural radiation are
generally not a problem since they are relatively constant. However, it
does vary geographically. Lead shielding around the counting chamber
eliminates much of the cosmic and natural radiation. Another potential source of background is
40
K found in glass. Low potassium glass, or plastic, should be used in the manufacture of
scintillation vials. The instrument is also a source of background radiation due to spontaneous
electrical pulses. Instrument background is controlled for by the coincidence circuits.

The most problematic source of background is often the sample itself. The nature of
sample background depends upon the experiment. Sample background is determined from
'blanks' which contain radioactivity but no sample. For example, if radioactivity is being used to
determine enzyme activity, then a blank containing all
components except the enzyme needs to be counted and
subtracted from the sample counts.

Statistics of Counting. Nuclear disintegrations are
random and infrequent. The accuracy of counting depends
cosmic
natural
machine
sample
68% confidence interval
counts (counts)

95% confidence interval
counts 1.96(counts)

40
upon total number of disintegrations observed (Box). The proportional error at a 95%
confidence level is defined by 1.96(counts)

/(counts). In other words you need ~10,000 total

counts to be 95% certain that the number determined is within 2% of the true value. In
practical terms this means that samples with low levels of radioactivity need to be counted
longer. For example, if you count a sample for 1 minute and it contains 1000 cpm (i.e., 1000
total counts) the error term is 6%. If you count the same sample for 10 minutes and it contains
1000 cpm (i.e., 10,000 total counts) the error term is now 2%.

Efficiency of counting. Not all disintegrations are detected and recorded as counts
during scintillation spectrophotometry. The efficiency of counting refers to the ratio of the cpm
to the dpm (i.e., cpm = dpm x efficiency). The inability to detect a disintegration can be due
to the light pulse being below the threshold of detection. In other words, the light intensity is too
low to result in an electrical signal. The efficiency of detecting radioactivity is generally directly
proportional to the energy level of the isotope.

Quenching will also affect the efficiency of counting. Quenching is the reduction in the
transfer of energy from the -particles to the photomultiplier tube. Chemical quenching is the
prevention of fluorescence of the fluors by substances in the sample (eg., strong acids or bases).
Color quenching is the absorption of the energy of the -particle or photon emitted by fluor or
solvent by substances in the sample (eg., chromophores). The efficiency of counting is not a
problem if the efficiency is the same for all the samples. However, if the efficiency (i.e.,
quenching) is different between the samples then it is necessary to determine the dpm in the
each of samples. Three methods by which efficiency can be determined are: 1) the internal
standard method, 2) the channels ratio method, and 3) the external channels ratio method.

1) internal standard

After the samples have been counted a known amount of
radioactivity (usually
3
H-toluene or
14
C-toluene) is added to each
sample and the samples are recounted. From this data the
efficiency can be determined for each sample (see Appendix 1
for calculations). The limitations of this method are that it is
tedious, expensive, and the sample cannot be recovered.

2) channels ratio

Quenching reduces the efficiency by shifting the energy
spectrum to a lower level. Different amounts of the quenching
agent are added to a known amount of radio-
activity and the cpm in two 'channels' are deter-
mined. A standard curve is generated by plotting
efficiency vs. the channel ratios (see Appendix 2).
The efficiency of each sample is then determined
by comparing the channel ratios of each sample to
the standard curve. Many scintillation counters
can be programmed to carry out this analysis and
41
to calculate the results. The channels ratio method is not very accurate in cases of low
radioactivity.

3) external channels ratio

An external radiation source (usually
137
Cs) is placed next to the sample and the
radioactivity determined. Electrons are ejected from solvent molecules by the -rays and behave
like -particles. This method incorporates principles of both internal standard and channels ratio
methods. The calculation is complex and carried out by the instrument.

Counting multiple isotopes. Scintillation spectrophotometry can also determine the
amount of different isotopes in a mixture. In the case of isotopes that have little overlap in their
energy spectrums, such as
3
H and
32
P, it is relatively easy. Channels are set so that one channel
contains low energy isotope and the other channel contains the high energy isotope. In cases
where the energy spectrums overlap it will be necessary to use channel ratios to determine
amount of each isotope. Radioactive standards are use to determine how much of each isotope
is found in each channel. If possible, the lower energy channel should be set so that it contains
all of the lower energy isotope. In other words, virtually no counts from the lower energy
isotope should be detected in the higher energy channel. (See Appendix 3 for sample
calculation.)

42
GAMMA-RAY DETECTION

Gamma-rays can be detected and
measured by the same methods as -particles
(eg., G-M counter, autoradiography, and
LSC). Like -particles, -rays are ionizing
radiation and will be detectable with a G-M
counter. Similarly, -rays are the same as X-
rays (i.e., photons) and will expose
photographic emulsions. In addition,
electrons (i.e., -particles) are ejected when
-rays interact with other substances.
Although it is possible to measure -rays
with a liquid scintillation counter, it is more common to use an instrument specifically designed
to count -radiation. The counting chamber of a -counter is surrounded by a thallium-activated
NaI crystal. When -rays interact with the NaI(T) secondary electrons (-particles) are ejected.
The -particles then result in radiation in the visible range (i.e., light) which is detected and
quantitated by photomultiplier tubes as in scintillation spectrophotometry. The advantage of the
-counter is that no scintillation cocktail is necessary.


SAFETY

Radioisotopes are a health hazard since ionizing radiation can lead to cellular damage
and DNA mutations. Radioisotopes need to be handled safely at all times and precautions
against exposure must always be taken. The time of exposure to radioactivity should be kept at
a minimum. Whenever possible plexiglass shielding should be placed between the user and the
radioactive sample when moderate and high energy isotopes are being used. Lab coats and
disposable gloves should be worn at all times. Avoid ingestion of isotopes by observing the no
eating, drinking, smoking, mouth-pipetting, etc. rules in the laboratory. Some users tend to be
more relaxed with the use of low energy isotopes such as
3
H than with higher energy isotopes
such as
32
P. However,
3
H incorporated into the cell can cause more damage than
32
P since
3
H
only travels a short distance and will have more of a chance to damage cellular structures.
Because of its high energy,
32
P will be more likely to exit the cell without doing any damage.

43
APPENDIX 1. INTERNAL STANDARDS

Calculation of efficiency using the internal standard method.

1. Measure the radioactivity in all of the samples (=cpm
1
).

2. Add a known amount of radioactivity (dpm
std
) to each of the samples and recount
(=cpm
2
).

3. Calculate the efficiency (E) for each sample using the following equation:

E = (cpm
2
- cpm
1
)/dpm
std
*

4. Divide the cpm for each of the samples (cpm
1
) by the efficiency (E) for that particular
sample resulting in the dpm
sample
.

*Derivation of equation:

1) cpm
2
= cpm
1
+ cpm
std
(The cpm after the addition of the radioactivity will be
equal to the original cpm + the cpm of the standard.)

2) E = cpm
std
/dpm
std
(This is the definition of Efficiency.)

3) cpm
std
= (E)dpm
std
(Rearrange eq. 2 and solve for cpm
std
.)

4) cpm
2
= cpm
1
+ (E)dpm
std
(Substitute the value of cpm
std
from eq. 3 into eq. 1.)

5) E = (cpm
2
- cpm
1
)/dpm
std
(Solve eq. 4 for Efficiency.)


44
APPENDIX 2. CHANNELS RATIO

Calculation of efficiency using the channels ratio method.

1. Take a known amount of radioactivity and add increasing amounts of the suspected
quenching agent and determine the cpm in two channels (A and B).

2. Calculate the efficiency for each of the standards containing the quenching agent using
eq. 2 from above.

3. Plot the channel ratios (cpm
B
/cpm
A
) on the y-axis vs. the efficiency on the x-axis to
make a standard curve like the figure below. (Can also use cpm
A
/cpm
B
as the channels
ratio.)

4. From the channels ratio of each sample determine the efficiency of each sample using
the standard curve.

5. Divide the total cpm (cpm
A
+ cpm
B
) in each sample by the efficiency for that sample to
determine the dpm for each sample.

45

APPENDIX 3. MULTIPLE ISOTOPES

Example for calculation of multiple isotopes in mixed samples.

1. Set channels so that virtually all of the counts of the lower energy isotope (eg.,
3
H) are
in channel A.

2. Determine the cpm in both channels A and B for all of the samples as well as a
3
H
standard and a
14
C standard. (Standards should be counted under same conditions as
samples. If samples exhibit different levels of quenching then the dpm for the samples
and standards will need to be determined.)

3. Use the channel ratios of the standards to calculate the amount of each isotope in each
channel. Sample data:

Radi oact i vi t y ( cpm) Channel Rat i os
Sampl e Channel A Channel B B/ A A/ B
Exper i ment al 500 800
3
H st andar d 16, 840 25 0. 0015 -
14
C st andar d 4250 10, 200 - 0. 416


3
H i n channel B: ( 0. 0015) ( 500) < 0. 75 cpm( si nce some of t he
500 cpmar e due t o
14
C, t her ef or e

14
C i n channel B 800 cpm*

14
C i n channel A = ( 0. 416) ( 800) = 333 cpm

3
H i n channel A = 500 - 333 = 167
Tot al
3
H = 167 cpm
Tot al
14
C = 800 + 333 = 1133 cpm

*I n ot her wor ds, at t hese channel set t i ngs t her e i s no
si gni f i cant amount of
3
H i n channel B and al l of t he
r adi oact i vi t y can be assumed t o be
14
C.

46
47

CHAPTER 5--pH AND BUFFERS

Maintaining the proper pH during experimental procedures is often one of the most
critical factors in determining whether an experiment will work. Buffers are the components in
solutions that maintain constant pH at a defined value. The pH is maintained due to an
equilibrium established between weak acids and bases. The same principles can also be applied
to other equilibrium phenomenon.

DEFINITIONS

An acid donates a proton (H
+
) by dissociating into its conjugate base:

acid conjugate base
-
+ H
+


A base accepts a proton (H
+
) forming a conjugate acid:

base + H
+
conjugate acid
+


A strong acid is one that completely ionizes in solution while weak acids only partially ionize
in solution. The weak acid and its conjugate base are in equilibrium. The equilibrium between
the acid and its conjugate base will determine the pH and weak acids are used as pH buffers. pH
is defined as the -log [H
+
] (where [] brackets refer to molar concentration).

HENDERSON-HASSELBACH

The ionization of weak acids is defined in part by the chemical properties of the
individual weak acid. Therefore, it is possible to calculate the pH of a solution containing a
weak acid and its conjugate base. The ionization of a weak acid can be defined by the following
equation:

HA A
-
+ H
+


The dissociation constant, or K
a
= [H
+
][A
-
]/[HA]. Solving for [H
+
] results in:

[H
+
] = [HA]K
a
/[A
-
]

Since pH = -log[H
+
], take the -log of both sides of the equation:

-log[H
+
] = (-logK
a
) + (-log[HA]/[A
-
])

Substitutions and rearrangements yield the Henderson-Hasselbach equation:

pH = pK
a
+ log[A
-
]/[HA]

The pK
a
is equal to the pH when the concentrations of the protonated acid and the ionized form
48
are equal. The pK
a
value is dependent on the particular weak acid or ionizable group and needs
to be either looked up or determined. The equation is also written as:

pH = pK
a
+ log[salt]/[acid]

The Henderson-Hasselbach equation can be used to calculate the ratio of acid and salt (=
conjugate base) that is needed to produce a buffer solution of desired pH. (See appendix for
examples.)

INSTRUMENTATION

The pH meter is composed of 1) a reference
electrode, 2) a glass electrode, and 3) a voltmeter. (The two
electrodes are usually combined into a single electrode.)
The glass electrode is made of a special glass that is
permeable to H
+
, but not other cations or anions. The
electrode is filled with 0.1 N HCl. The H
+
diffuses from
higher concentration to lower concentration leaving behind
Cl
-
, thus creating an electrical potential. This electrical
potential is proportional to the pH of the solution being
measured as expressed by:

Voltage = E
o
+ (2.303RT/F)pH

where E
o
= summation of electrical potentials (i.e., voltage)
that are not pH related, R = gas constant, T = absolute temperature, F = Faraday constant. A
voltmeter will measure this electrical potential of the glass electrode (i.e., E) as compared to the
reference electrode (i.e., E
o
) and convert it to a pH value. It is necessary to calibrate the meter
using a solution of known pH before use due to changes in electrode HCl concentration and
other factors that affect E
o
.

COMPLICATIONS

Several factors can affect pH and/or the measurement of
pH (Box). Generally, these factors will have a minimal effect. In
addition, some electrodes are incompatible with certain buffers; the most notable being Tris
buffers.

Ionic Concentration. The concentration of ions and in particular the concentration of the
buffer itself can affect pH. Although Henderson-Hasselbach is independent of concentration the
equation is most accurate at low buffer concentration. This is especially important when stock
solutions that will be diluted to the final buffer concentration are used. Therefore it is generally
best to make the final pH adjustment at the time the buffer is diluted.

Na
+
Error. Sometimes pH electrodes will read the [Na
+
] in addition to the [H
+
]. This
occurs most often at high pH; therefore it is better to use KOH to adjust the pH of alkaline
ionic concentration
Na
+
error
temperature
49
buffers.

Temperature. The temperature affects the pK
a
of a buffer. The degree that this occurs is
dependent on the buffer. For example, Tris has a pK
a
/C
o
= -0.03. Therefore, it is important to
adjust the pH at the temperature at which the buffer will be used.

BUFFERING CAPACITY

Buffering capacity refers to the amount of acid or base you can add to the system
without greatly affecting the pH. Buffering capacity () is defined in operational terms:

= B/pH

where B = the mole equivalent of strong base or strong acid added and pH is the resulting
change in pH. Buffering capacity will be the greatest at the pK
a
of the buffer.

BUFFER SELECTION

There are no well defined criteria for
choosing a buffer except that the pK
a
should be
close to the desired pH range ( 1 pH unit) since
the buffering capacity is greatest at the pK
a
.
Buffers should not participate in reactions that
are being analyzed or otherwise affect the out-
come of the experiment. For example, the buffer
should not bind to other components in the
mixture. The most common criteria for choosing
a buffer is the scientific literature. In other
words, use the same buffer for a particular
procedure that has worked before. The particular
experiment also needs to be taken into account. For example, if a reaction produces H
+
, then the
buffer should have a pK
a
lower than the desired pH. This will provide more buffering capacity
since the pH will move towards the pK
a
as the reaction proceeds.

During the 1960's Dr. Norman Good examined a series of zwitterionic compounds as
biological buffers based upon some ideal properties (Box). A series of such buffers--often
called Good buffers--with pK
a
values ranging from 6-10 was developed. These buffers have
long names and are more commonly known by their abbreviations (eg., HEPES).

Goods Buffer Criteria
pK
a
between 6 and 8
high solubility in aqueous solutions
exclusion by biological membranes
minimal salt effects
minimal effects on dissociation due
to concentration, temperature and
ionic composition
well-defined or non-existent
interactions with cations
chemical stability
insignificant light absorption
between 240 and 700 nm
readily available in pure form
50
BUFFER APPENDIX

Example 1. How many grams of acetic acid (CH
3
CO
2
H) and sodium acetate are needed to make
up one liter of a 50 mM acetate buffer with a pH of 5.0?

5.0 = 4.76 + log[Naacetate]/[acetic a.] Substitute values into Henderson-Hasselbach

0.24 = log[Naacetate]/[acetic a.] Subtract 4.76 from both sides.

1.74 = [Naacetate]/[acetic a.] Take anti-log (10
X
) of both sides

[Naacetate] = 1.74[acetic a.] Multiply both sides by [acetic a.] (Eq. 1)

[Naacetate] + [acetic a.] = 0.05 The sodium acetate and acetic acid
concentrations must add up to the desire final
molar concentration of the buffer. (Eq. 2)

1.74[acetic a.] + [acetic a.] = 0.05 Substitute eq. 1 into eq. 2.

2.74[acetic a.] = 0.05

[acetic a.] = 0.018 moles/liter Eq. 3

[Naacetate] + 0.018 = 0.05 Substitute eq. 3 into eq. 2

[Naacetate] = 0.032 moles/liter Eq. 4

0.018 moles/liter x 60 g/mole x 1 liter = 1.09 grams of acetic acid

0.032 moles/liter x 82 g/mole x 1 liter = 2.62 grams of sodium acetate

Example 2. After making the calculations for example 1, you realize that there is no available
sodium acetate. However, you do have a 1 M NaOH solution. How is the buffer made now?

The total acetate concentration must now come from the acetic acid, so:

0.05 moles/liter x 60 g/mole x 1 liter = 3.0 grams of acetic acid

The calculations from example 1 are still valid. The only change is to substitute [NaOH] for
[Naacetate] in eq. 4. Concentrations are easily calculated using:

C
1
V
1
= C
2
V
2


Therefore (0.032M)(1 liter) = (1M)V
2
, so V
2
= 32 ml of 1 N NaOH

51
Example 3. How many grams of NaH
2
PO
4
and Na
2
HPO
4
are needed to make 50 ml of a 0.1 M
phosphate buffer with a pH of 7.2? Phosphate has three ionizable groups with different pK
a

values. Choose the pK
a
closest to the desired pH. Since pH = pK
a
use equal molar amounts of
the monobasic and dibasic phosphates.

NaH
2
PO
4
= 0.05 moles/liter x 138 g/mole x 0.05 liter = 0.345 g

Na
2
HPO
4
= 0.05 moles/liter x 142 g/mole x 0.05 liter = 0.355 g

Example 4. How many grams of Tris and TrisHCl are needed to make up 500 ml of 1 M Tris
buffer with a pH of 8.0? Although Tris is a base the calculations are the same. Make sure that
the base form and the conjugate acid form are correctly identified.

8.0 = 8.3 + log[Tris]/[TrisHCl]

[Tris] = 0.5[TrisHCl] and [Tris] + [TrisHCl] = 1

[TrisHCl] = 0.67 M x 157.6 g/mole x 0.5 l = 52.8 g

[Tris] = 0.33 M x 121 g/mole x 0.5 l = 20 g

ADDITIONAL PRACTICE PROBLEMS

How many grams of acetic acid and how many ml of 1
M NaOH are needed to make 1 liter of 50 mM acetate
buffer, pH 5.0? The pK
a
of acetate is 4.76 and
molecular weight of acetic acid is 60.

How many grams of HEPES and how many ml of 1 M
NaOH are needed to prepare 1 liter of a 0.1 M HEPES
buffer at pH 7.4? The pK
a
of HEPES is 7.5 and the
molecular weight is 238.

How many grams of each Tris and TrisHCl are need to
make up 1 liter of 0.5 M Tris buffer at pH 8.8? The
pK
a
of Tris is 8.3 and the molecular weights of Tris
and TrisHCl are 121 and 157 respectively.

How many grams of MES (mol. wt. = 195) and the Na
+
salt of MES (mol. wt. = 217) are
needed to make 100 ml of 0.5 M MES buffer at pH 6.0. pK
a
of MES = 6.15?

How many grams of Tricine (mol. wt. = 179) and how many ml of 1 N HCl are needed to make
of 200 ml of 0.1 M Tricine buffer at pH 8.5. The pK
a
of Tricine = 8.15?
Buffer Mol. Wt. pK
a

acetate 60 (free acid)
82 (Na salt)
4.76
phosphate 138 (monobasic)
142 (dibasic)
2.15
7.2
12.3
MES 195 (free acid)
212 (Na salt)
6.15
HEPES 238 (free acid) 7.5
TRIS 121 (free base)
157 (TrisHCl)
8.3
Tricine 179 (free base) 8.15
52


53
CHAPTER 6--CENTRIFUGATION

The centrifuge is an essential instrument in cell and
molecular biology research. It is primarily used to separated biolo-
gical components based upon differential sedimentation properties.
Many types of centrifuges are available for various applications
(Box). All centrifuges basically consist of a motor which spins a
rotor containing the experimental sample. The differences between
centrifuges are in the speeds at which the samples are centrifuged
and the volumes of samples.

SEDIMENTATION THEORY

Spinning around an axis creates a
centrifugal field (Figure). A particle in a
centrifugal field will experience a centrifugal
force defined by:

F
c
= m
2
r

where F
c
= the centrifugal force, m = mass of
the particle, = angular velocity and r =
distance from the axis. This force will be
opposed by a buoyant force (F
b
) and a frictional force (F
f
).The buoyant force represents the
force it takes to displace solvent as the particle moves through the centrifugal field. The
frictional force represents the drag on the particle as it passes solvent molecules. These two
forces are respectively defined as:

F
b
= -m
o

2
r and F
f
= -fv

where m
o
= the mass of the displaced solution, f = frictional coefficient and v = velocity of the
particle. The particle will move at a velocity such that the total force equals 0, thus:

F
c
+ F
b
+ F
f
= 0, or m
2
r - m
o

2
r - fv = 0

substituting m
s
= m
o
, where = partial specific volume of the particle and
s
= density of the
solvent, and solving for v results in:

v =
2
rm(1 -
s
)/f =
2
rm(
p
-
s
)/f

This equation (expressed in the two different forms) tells us several things about sedimentation:

1. The more massive a particle, the faster it moves in a centrifugal field

2. The denser a particle (i.e., the smaller its ) the faster in moves in a centrifugal field.

Ultracentrifuge
Analytical
Preparative
High Speed
Table Top
Clinical
Microfuges
Centrifugal Force
54
3. The denser the solution, the slower the particle will move in a centrifugal field.

4. The greater the frictional coefficient (factors such as viscosity, particle shape, etc.
influence this parameter), the slower the particle will move.

5. The particle velocity is 0 when the solution density is greater than the particle density.

6. The greater the centrifugal force (
2
r) the faster the particle sediments.

The velocity per unit force will be defined as the sedimentation coefficient (s), or:

s = v/
2
r = m(1 -
s
)/f

When mass is expressed in g and f in g/sec s ranges from 10
-13
to 10
-11
sec. This is normally
expressed in Svedberg (S) units where 1 S = 10
-13
sec. The higher the S value, the faster that
particle will sediment. For example, subcellular compartments and macromolecules have
sedimentation coefficients which reflect their size, shape and density (Figure).



The sedimentation coefficient is determined by measure the velocity of a particle in a
known centrifugal field. S units need to be normalized according to the composition of the
medium. Sedimentation coefficients are usually determined with an analytical ultracen-
trifugation, also referred to as a model E ultracentrifuge. The rotor in the model E centrifuge
allows the sample to be monitored spectrophotometrically as the sample is centrifuging. As
the sample migrates in the centrifugal field the absorbance across the chamber will change.
The sedimentation rate can be calculated from this change in absorbance. In addition,
physical characteristics (such as size, density and shape) of a particle, or molecule, can be
55
determined from the sedimentation rate of a particle in a
medium of known composition.

PREPARATIVE CENTRIFUGATION

The centrifuge is used most often in the biological
sciences to collect material or to separate particles based upon
their sedimentation properties. Preparative centrifugation
takes advantage of the fact that more massive particles will
sediment faster than less massive particles. For example,
organelles and other subcellular components can be isolated
by differential centrifugation (Figure). Differential centrifuga-
tion is carried out by centrifuging a sample at low speed and
separating the supernatant and pellet. The supernatant is then
recentrifuged at higher speed and the supernatant and pellet
separated again.

Sedimentation of a particular particle is dependent
upon the RCF (see box) and how long the centrifugal force is applied. Samples must be
centrifuged long enough for the particles at the top of the tube to reach the bottom. The size and
shape of the centrifuge tube will also affect the centrifugation time. In addition, there is a
difference in the RCF between the top and bottom of the tube. These factors need to be taken
into consideration when scaling up from small pilot experiments carried out in small centrifuge
tubes into large scale experiments. The term gmin refers to the product of the minutes a sample
is centrifuged and the g-force. In general, assuming that the sizes and dimensions of the
centrifuge tubes are similar, the product of the g-force and the time of centrifugation can be
used to determine the conditions needed to to completely sediment a particle. For example, if it
takes 60 min to completely sediment a particle at 1000 x g, then it should only take 10 min to
sediment the same particle at 6000 x g.

Relative centrifugal force (RCF) is defined as the
ratio of the centrifugal force to the force of gravity:

RCF = F
c
/F
g
=
2
r/980

(expressed in radians/sec) is converted to
revolutions per minute (rpm) by substituting =
(rpm)/30, resulting in:

RCF = 1.119 x 10
-5
(rpm)
2
r

where r is expressed in cm. RCF units are expressed
as "x g". (See appendix for nonograph.)
56
Centrifugation Through Density Gradients

Fast sedimenting particles will be contaminated with slow
sedimenting particles. The reason for this is that by the time large particles
near the top of the tube are pelleted, some of the small particles near the
bottom of the tube have also pelleted. In addition, mechanical vibrations,
thermal gradients and convection currents can also affect the sedimentation
properties. Centrifugation through a dense medium, or density gradient centrifugation, can
partially alleviate these problems. In addition, density gradient centrifugation will allow for
better separation of particles with similar properties. Several different media are commonly
used in density gradient centrifugation depending upon the exact application (Box).

The two types of density gradient centrifugation are rate zonal and isopycnic (or
equilibrium). In rate zonal centrifugation the density of the particles being separated are greater
than the density of the solvent. Separation is based primarily upon size (i.e., larger particles will
sediment faster). It is important to determine the optimal length of centrifugation for separating
the particle of interest. If the centrifuge is not turned off soon enough all of particles will pellet.
In isopycnic centrifugation the solvent density encompasses density of particles. The separation
is based upon particle density. Centrifugation is carried out until equilibrium is reached (i.e., all
particles have banded at densities corresponding to their own).

Density gradients can be preformed or formed during centrifugation. In the case of
isopycnic density gradients it is possible to mix the sample with the desity gradient medium and
carry out the centrifugation until the density gradient forms. The various components in the
sample will then be found at a position which corresponds to their density. This is especially
useful in the case of Percoll which rapidly forms density gradients when subjected to
centrifugation. Another common example in which self-forming gradients are commonly used
is the separation of nucleic acids on CsCl gradients (see Nucleic Acid Struction and Isolation).



Preforming the gradients before centrifugation decreases the amont of time and centri-
fugal force needed to reach equilibrium. In the case of rate zonal density gradients it is
necessary to use preformed gradients. Gradient makers are used to produce continuous
gradients, or step-wise gradients can be prepared by layering successive solutions of lesser
density. Samples are layered onto the gradients and subjected to centrifugation. In the case of
isopycnic gradients the sample can also be underlaid at the bottom of the tube and the various
particles will 'float' to their correct densities during centrifugation. Following centrifugation the
gradient is divided into fractions corresponding to different densities and analyzed.
Sucrose
CsCl
Ficoll
Hypaque
Percoll
57

The fractions are analyzed for the
component(s) of interest and if needed the
densities of the various frations can be
determined. One method of determining
the density is to measure the refractive
index of the fractions and to then calculate
the densities based on a standard curve
prepared from solutions of known density
made with the same medium of the desity
gradients (eg., sucrose, etc.). Another
method of determining the density is to
prepare an identical gradient with 'marker
beads' of known density. These marker beads are different colors and the position in the
gradient is determined by measuring the distance from the top of the tube or determining
which fraction they are associated with. This can then be use to estimate the range of
densities in the other tubes containing sample. In subcellular fractionation experiments the
various organelles can be evaluated by measuring components (eg., enzymes) known to be
exclusively associated with the various subcellular compartments. This will identify the
fraction(s) containing the organelle of interest as well as provide information about
contamination with other organelles.
Subcellular Fractionation
and Marker Enzymes
Organelle Marker
nuclei DNA
mitochondria cytochrome oxidase
lysosome hydrolases
peroxisome catalase
Golgi -mannosidase
plasma membrane adenylate cyclase
cytosol lactate dehydrogenase
58
CENTRIFUGATION APPENDIX 1. RCF CALCULATION

The relative centrifugal force (RCF) can be calculated from the following equation:

RCF = (1.119 x 10
-5
)(rpm)
2
(r)

where rpm is the speed of rotation expressed in revolutions per minute and r (radius) is the
distance from the axis expressed in cm. The RCF units are "x g" where g represents the force of
gravity. RCF can also be determined from the nomograph below. Place a straight edge to
intersect the radius and the desired RCF to calculate the needed rpm. Alternatively place the
straight edge on the radius and the rpm to calculate the g-force. For example, spinning a sample
at 2500 rpm in a rotor with a 7.7 cm radius results in a RCF of 550 x g.


59



PART II



Analysis and Characterization of
Proteins



Topics covered:

Protein Structure and Assays
Differential Solubility
Chromatography
Membranes and Detergents
Electrophoresis
Overview on Protein Purification


61
CHAPTER 7--INTRODUCTION TO PROTEINS

Proteins typically make up more than half the dry weight of cells. They contribute to the
structure of a cell and are responsible for cellular functions such as catalysis and molecular
recognition.

PROTEIN STRUCTURE

Proteins are polymers of L--amino acids. The refers
to a carbon with a primary amine, a carboxylic acid, a hydrogen
and a variable side-chain group, usually designated as 'R'.
Carbon atoms with four different groups are asymmetric and
can exhibit two different arrangements in space due to the
tetrahedral nature of the bonds. The L refers to one of these two
possible configurations the four different groups on the -
carbon can exhibit. Amino acids of the D-configuration are not
found in proteins and do not participate in biological reactions.

Twenty different amino acids, distinguished by their side-chain
groups, are found in proteins (Box). The side-chain groups vary
in terms of their chemical properties such as polarity, charge and size. These various side-chain
groups will influence the chemical properties of proteins as well as determine the overall
structure of the protein (see Appendix). For example, the polar amino acids tend to be on the
outside of the protein where they interact with water and the nonpolar amino acids are on the
inside forming a hydrophobic core.

The covalent linkage between two amino acids
is known as a peptide bond. A peptide bond is formed
when the amino group of one amino acid condenses
with the carboxyl group of another amino acid to form
an amide (Figure). This arrangement gives the
polypeptide chain a polarity in that one end will have a
free amino group, called the N-terminus, and the other
end will have a free carboxyl group, called the C-
terminus.

Peptide bonds tend to be planar which gives
the polypeptide backbone some rigidity. However,
rotation can occur around both of the -carbon bonds
resulting in a polypeptide backbone with different potential conformations in regards to the
relative positions of the R-groups. (Conceptually this can be viewed as the R-groups projection
into or out from the page in the figure.) Although many conformations are theoretically
possible, interactions between the R-groups will limit the number of potential conformations
and proteins tend to only fold into a single functional conformation. In other words, the confor-
mation or shape of the protein is due to the interactions of the side-chain groups with one
another and with the polypeptide backbone. The interactions can be between amino acids that
L--AMINO ACIDS

Nonpolar Polar

Alanine Arginine
Glycine Asparagine
Isoleucine Aspartic a.
Leucine Cysteine
Methionine Glutamic a.
Phenylalanine Glutamine
Proline Histidine
Tryptophan Lysine
Valine Serine
Threonine
Tyrosine
62
are close together in a polypeptide or between amino acids that are far apart or even on
different polypeptides. These different types of interactions are often discussed in terms of
primary, secondary, tertiary and quaternary protein structure (Table).

Levels of Protein Structure
Primary
Refers to the amino acid sequence and the location of disulfide
bonds between cysteine residues (i.e., covalent bonds).
Secondary
Refers to interactions between amino acids that are close
together (eg., -helix, -sheet, -turn, random coil).
Tertiary
Refers to interactions between amino acids that are far apart
(eg., motifs, domains).
Quaternary
Refers to interactions between two or more polypeptide chains
(i.e., protein subunits).

The primary amino acid sequence and positions of disulfide bonds strongly influence
the overall structure of protein. In regarads to the primary amino acid sequence, certain side-
chains will permit, or promote, hydrogen-bonding between neighboring amino acids of the
polypeptide backbone resulting in secondary structures such as -sheets or -helices. Alterna-
tively, certain R-groups may interfere with each other and prevent certain conformations.
In the -helix conformation the peptide backbone takes on a 'sprial staircase' shape
which is stabilized by H-bonds between carbonyl and amide groups of every fourth amino acid
residue. This restricts the rotation of the bonds in the peptide backbone resulting in a rigid
structure. -sheets are also rigid structures in which the polypeptide chain is nearly fully
extended with the R-groups alternating between pointing up and down. -sheets interact either
in parallel (both with same orientation in regards to N- and C-termini) or anti-parallel fashion.
Certain amino acids promote the formation of either -helices or -sheets due to the nature of
the side-chain groups. Some side-chain groups may prevent the formation of secondary
structures and result in a more flexible polypeptide backbone, which is often called random coil
conformation.
The other aspect of primary protein
structure is the position of disulfide bonds. The
amino acid cysteine has a free thiol group that
can be oxidized to form a covalent bond with
another cysteine (Figure). These disulfide bonds
can form between cysteine residues that are relatively close or far apart within a single polypep-
tide chain, or even between separate polypeptide subunits with a protein. In this regard,
disulfide bonds can contribute to secondary, tertiary and quaternary aspects of protein structure.
Proteins containing disulfide bonds will be sensitive to reducing agents (such as -mercapto-
ethanol) which can break the disulfide bond.
The various secondary structures can interact with other secondary structures within the
same polypeptide to form motifs or domains (i.e., tertiary structure). A motif is a common
combination of secondary structures and a domain is a portion of a protein that folds
independently. The tertiary structure will represent the overall three dimensional shape of a
63
polypeptide. A typical protein structure is a compact entity composed of the various secondary
structural elements with protruding loops of
flexible (ie, random coil) sequence. This is often
depicted in a ribbon diagram (Figure) in which -
sheets are drawn as flat arrows with the arrowhead
representing the N-terminal side and -helices are
drawn as flat spirals. The flexible loops are repre-
sented by the strings connecting the secondary
structural elements.
Many proteins are composed of multiple subunits,
or distinct polypeptide chain that interact with one
another. This is referred to as quaternary structure.

PROTEIN STABILITY

Proteins are often fragile molecules
that need to be protected during purifi-
cation and characterization. Protein dena-
turation refers the loss of protein structure
due to unfolding. Maintaining biological
activity is often important and protein
denaturation should be avoided in those
situations. Elevated temperatures, extremes in pH, and changes in chemical or physical
environment can all lead to protein denaturation (Table). In general, things that destabilize H-
bonding and other forces that contribute to secondary and tertiary protein structure will promote
protein denaturation. Different proteins exhibit different degrees of sensitivity to denaturing
agents and some proteins can be re-folded to their correct conformations following
denaturation.

Factors Affecting Protein Stability
Factor Possible Remedies
temperature Avoid high temperatures. Keep solutions on ice.
freeze-thaw
Determine effects of freezing. Include glycerol in buffers. Store in
aliquots.
physical
denaturation
Do not shake, vortex or stir vigorously. (Protein solutions should
not foam.)
solution effects Mimic cellular environment: neutral pH, ionic composition, etc.
dilution effects Maintain protein concentrations > 1 mg/ml as much as possible.
oxidation Include 0.1-1 mM DTT (or -ME) in buffers.
heavy metals Include 1-10 mM EDTA in buffers.
microbial growth Use sterile solutions, include anti-microbials, and/or freeze.
proteases Include protease inhibitors. Keep on ice.

The optimal conditions for maintaining the stability of each individual protein need to
64
be determined empirically. In general, though, protein solutions should be kept cold (< 4
o
C)
except during assays and other procedures requiring specific temperatures. Many proteins are
especially labile and need to be stored at -20
o
or -80
o
. However, repeated freezing and thawing
of protein solutions is often deleterious. Adding 50% glycerol to storage buffers will lower the
freezing point and allow storage at -20
o
. Solutions for working with proteins will often contain
heavy-metal chelators and/or antioxidants as protectants.. In addition, proteases may be released
during cell disruption and it may therefore be necessary to include protease inhibitors.


PROTEIN ASSAYS

Numerous spectrophotometric methods have been developed to estimate the amount of
protein in a sample. Proteins are chromophores with absorption maximum in the UV range.
Some proteins, such as cytochromes and hemoglobin, will have distinct spectral characteristics
due to prosthetic groups. In addition, several indirect ways to measure protein concentrations
spectrophotometrically have been developed.

UV Absorption

A simple method to measure protein concentration is to determine the absorption at 280
nm. Tyrosine and tryptophan residues which have A
max
at 275 and 280, respectively, are
responsible for this absorption. The distribution of tyrosine and tryptophan is fairly constant
among proteins so it is not absolutely necessary to determine an extinction coefficient for each
individual protein. Typically, a 1 mg/ml protein solution will result in an A
280
of approximately
one (1). In the case of purified proteins the exact extinction coefficient will depend on the exact
amount of tyrosine and tryptophan in that particular protein. For example, a 1 mg/ml IgG
solution has an A
280
of approximately 1.5. The simplicity and ability to completely recover the
sample are the major advantages of this method. The lower limit of sensitivity for UV
absorption is 5-10 g/ml.

A potential problem with using A
280
values to calculate protein concentration is the
absorption due to contaminating substances, and in particular, nucleic acids which have an A
max

at 260 nm. It is possible to use correction factors that permit the determination of protein
concentrations. A particularly convenient formula is:

(A
235
- A
280
)/2.51 = mg/ml protein.

Indirect spectrophometric assays (eg., Lowry, Bradford) for the determination of protein
concentrations overcome some of the problems associated with interfering substances in protein
samples. However, the measured protein cannot be recovered in such assays and they take
longer to perform.

Folin-Ciocalteu or Lowry

Historically, one of the most widely used protein assays was the Lowry assay. This
assay is a modification of a previous assay known as the Biuret. In the Lowry assay proteins
65
react with alkaline Cu
2+
reducing it to Cu
+
. The reduced Cu
+
and the side-chain (R) groups of
tryptophan, tyrosine and cysteine react with the Folin-Ciocalteu reagent (complex of inorganic
salts) to form a blue color that is proportional to the amount of protein. The A
600-750
is
determined and protein concentration is calculated from a standard curve. The assay is linear 1-
300 g and the lower limit of sensitivity is 1-5 g/ml. Substances in the sample such as
detergents can interfere with the results and therefore appropriate controls and blanks need to be
carried out.

Bradford or Coomassie Blue G-250

The Bradford assay has replaced the Lowry as the standard protein assay. The major
advantage is that it is carried out in a single step and that there are very few interfering
substances. The principle of the assay is based on a shift of the A
max
of the Coomassie-blue (G-
250) dye from 465 nm to 595 nm in the presence of protein due to a stabilization of the anionic
form of the dye. The dye reacts primarily with arginine residues and to lesser extent with his,
lysine, tyrosine, tryptophan and phenylalanine. Protein concentrations are determined by
developing a standard curve with known amounts of proteins and extrapolating the absorbance
values of the samples. The standard curves are not linear over a wide range of protein
concentrations. This assay can also carried out in 96-well plates and read on automated ELISA
readers. Programs are available that will automatically calculate the protein concentrations
based upon a standard curve.

Assay Of Specific Proteins

In addition to measuring the total amount of protein, it is
often necessary to estimate the amount of a specific protein in a
mixture of proteins. Measuring a specific protein will depend upon
the availability of an assay that is specific for the protein of interest.
Protein assays should be practical in addition to being specific and accurate (Box). Typically
protein assays are based upon the biological activity of the protein of interest. For example,
enzyme assays will detect the conversion of a substrate to a product. Enzymes assays can be
based upon colorimetric, fluorescent or radioactive substrates (or products). Many proteins bind
to ligands or other substances and this binding activity is measured. Bioassays measure a
change in some biological property (eg., stimulation of cell division). In cases where the protein
of interest has no measurable activity or the activity is unknown it may be possible to generate
antibodies against the protein and develop an immunoassay (to be discussed later). If antibodies
against such a protein are not available, the assay may simply be the amount of a protein band
on a Commassie blue-stained gel following electrophoresis (to be discussed later).


Specificity
Sensitivity
Accuracy
(Quantification)
Rapid
Easy to Perform
66
APPENDIX. AMINO ACIDS: PROPERTIES AND STRUCTURE

Proteins are composed of 20 different amino acids which are
distinguished by their side-chain groups, designated as R (see figure at
right). The amino acids are grouped below according to the chemical
nature of their side-chain groups.


Name/Abbreviations R-group Comments
Glycine Gly G -H
The lack of a side chain provides for the
greatest possible conformational
flexibility.
Proline Pro P

The aliphatic side-chain forms a cyclic
imino compound (entire amino acid is
shown). This imposes rigid constraints on
the rotation around the backbone and has
significant affects on protein conformation.
Amino Acids with Aliphatic Side Chains
Alanine Ala A -CH
3

Valine Val V

Leucine Leu L

Isoleucine Ile I

These alkyl side-chain groups are
important for hydrophobic interactions and
provide for a variety of surfaces and
shapes.
Amino Acids with Aromatic Side Chains
Phenylalanine Phe F

Tyrosine Tyr Y

Tryptophan Trp W

The aromatic groups are important for
hydrophobic interactions and may be
especially important for interacting with
other flat molecules.
Amino Acid Alcohols
Serine Ser S -CH
2
OH
Threonine Thr T

The hydroxyl groups are weakly ionizable
(pK
a
~13) and participate as active groups
in some enyzmes.
Acidic Amino Acids
Aspartatic acid Asp D

Glutamatic
acid
Glu E -CH
2
CH
2
CO
2
H
These side-chain groups are generally
negatively charged at neutral pH (pK
a
=
4.3-4.7).
67

Amides of the Acidic Amino Acids
Asparagine Asn N

Glutamine Gln Q -CH
2
CH
2
CONH
2
These side-chain groups do not ionize, but
are relatively polar.
Basic Amino Acids
Histidine His H

The imidazole group participates in the
active site of many enzymes, as well as
binding metal ions.
Lysine Lys K -CH
2
CH
2
CH
2
CH
2
NH
2
Arginine Arg R

These side-chain groups are generally
positively charged at neutal pH (pK
a
>10).
Sulfur Containing Amino Acids
Cysteine Cys C
-CH
2
SH
-CH
2
S-
SCH
2
-
Cysteines participate in redox reactions
and can form disulfide links between two
residues (i.e., oxidized state).
Methionine Met M -CH
2
CH
2
SCH
3

Met is rather hydrophobic, but the
thioether group is a potent nucleophile.
Other Abbreviations
Asp or Asn Asx B
Glu or Gln Glx Z
The amides are converted to the acids by the procedures used to
identify and quantify amino acid residues. Therefore they are
designated as either in such situations..
Any amino
acid
Xaa X Used when the amino acid is unknown or does not matter.


69
CHAPTER 8--DIFFERENTIAL PRECIPITATION OF PROTEINS

Proteins are responsible for many aspects of cell function and structure. To learn more
about their precise roles in cellular biology it is often necessary to analyze purified proteins. A
protein of interest can be separated from other proteins based upon its unique physical and
chemical properties. The first step in protein isolation is to extract the protein from the tissue or
cell in a soluble form. In some cases it is desirable to first disrupt the cell and to isolate the
appropriate subcellular compartment before solubilizing the protein. Generally, cytosolic
proteins are found in the supernatant following cell disruption and centrifugation at 100,000 xg
for 30-60 min. Membrane proteins or proteins associated with organelles will require additional
solubilization techniques such as detergents or chaotropic agents (see Chapter 10).

Proteins are soluble in aqueous solutions because the charged and polar side-chain
groups of the amino acids interact with water. In other words, proteins become hydrated. If the
protein-solvent interaction is prevented, proteins will interact with one another and form
aggregates that precipitate out of solution. Different proteins will behave differently in terms of
interacting with either the solvent or other protein molecules. Therefore it is possible to separate
different proteins based upon different solubility properties. For example, as the salt
concentration of a solution is increased the amount of water that is available to interact with
proteins is decreased. This will result in the more hydrophobic domains of proteins interacting
with hydrophobic domains of other proteins. The interacting proteins will form large masses, or
a precipitate, that can be collected by centrifugation. Proteins with a larger proportion of
hydrophobic amino acid residues exposed on their surface will be more sensitive to effects of
salts than highly charged proteins. Therefore, it is possible to selectively precipitate some
proteins under conditions where other proteins remain soluble. One of the most common salts
used for this differential precipitation of proteins is (NH
4
)
2
SO
4
.

PROCEDURE FOR (NH
4
)
2
SO
4
PRECIPITATION.

A common method for differentially precipitating proteins is by 'salting-out' with
(NH
4
)
2
SO
4
. As the (NH
4
)
2
SO
4
is added, H
2
O molecules interact with the salt ions, thus
decreasing the amount of water available to bind to protein. Proteins are differentially precipi-
tated by (NH
4
)
2
SO
4
due to their differences in hydrophobicity. A more hydrophobic protein will
precipitate at a lower salt concentration than a more hydrophilic protein.

1. The (NH
4
)
2
SO
4
is slowly added to a protein solution while gently stirring. The amount to
add to precipitate the protein of interest will have to be initially determined empirically. The
concentration of (NH
4
)
2
SO
4
needed to precipitate a particular protein in usually expressed
as a percent of saturated (NH
4
)
2
SO
4
. Either solid or a saturated solution can be added. For
large volumes, adding solid is more convenient.

2. Once dissolved, the (NH
4
)
2
SO
4
/protein suspension is allowed to slowly stir (usually while
kept cold) until equilibrium is reached. This is typically for one hour, however, some
proteins require overnight incubations. The precipitated protein are collected by centri-
fugation. Typically 10,000 xg for 10 minutes is sufficient to pellet precipitated proteins.
70

3. The protein pellet is dissolved in the buffer of interest and the remaining (NH
4
)
2
SO
4
is
removed by either dialysis or desalting column. Dialysis tubing has defined pore size which
allows small molecules to pass through freely while retaining larger molecules. Multiple
buffer changes accelerates the process.

(NH
4
)
2
SO
4
precipitation can also be
carried out as a two-step procedure. The maxi-
mum amount of (NH
4
)
2
SO
4
that does not precip-
itate the protein of interest is added and this
pellet is discarded. (NH
4
)
2
SO
4
is then added to
the supernatant to bring the concentration to the
minimal % saturation that completely precipi-
tates the protein of interest. Nomograms for
calculating the amounts of (NH
4
)
2
SO
4
needed to
achieve the desired percent saturation are avail-
able (see appendix). The pellet is retained and
the supernatant discarded at this step.

Another variation of (NH
4
)
2
SO
4
precipitation is 'back extraction'. In this case the
total protein is precipitated with 80-90% saturated (NH
4
)
2
SO
4
and the pellet is extracted
with the appropriate (NH
4
)
2
SO
4
concentration to solubilize the protein of interest.


OTHER PROTEIN PRECIPITATION METHODS

Organic solvents can also be used to differentially
precipitate proteins. Water miscible solvents (eg., acetone,
ethanol, methanol) are mixed with an aqueous solution. As the water molecules interact with
the organic solvent molecules less water is available for hydration and proteins will
differentially precipitate as in the case of added salts.

Proteins also tend to precipitate out at high temperatures or at extreme pH values.
Proteins will differentially denature under conditions of high temperature or extreme pH. The
unfolded, or denatured, proteins will have hydrophobic residues exposed and will aggregate.
Most of the time precipitation with high temperatures or acidic pH results in precipitated
protein which can not be resolubilized. However, it is possible to remove unwanted proteins in
cases where the protein of interest is acid or heat stable.
salt (eg, (NH
4
)
2
SO
4

solvents
acidic pH
high temperature
71
APPENDIX. AMMONIUM SULFATE NOMOGRAM

Final Concentration of Ammonium Sulfate (% saturation)
20 25 30 35 40 45 50 55 60 65 70 75 80 90

Grams of solid (NH
4
)
2
SO
4
to be added to 1 liter of solution.
0 114 144 176 209 243 277 313 351 390 430 472 516 561 662
20 29 59 91 123 155 189 225 262 300 340 382 424 520
25 30 61 93 125 158 193 230 267 307 348 390 485
30 30 62 94 127 162 198 235 273 314 356 449
35 31 63 94 129 164 200 238 278 319 411
40 31 63 97 132 168 205 245 285 375
45 32 65 99 134 171 210 250 339
50 33 66 101 137 176 214 302
55 33 67 103 141 179 264
60 34 69 105 143 227
65 34 70 107 190
70 35 72 153
75 36 115
I
n
i
t
i
a
l

C
o
n
c
e
n
t
r
a
t
i
o
n

o
f

A
m
m
o
n
i
u
m

S
u
l
f
a
t
e

(
%

s
a
t
u
r
a
t
i
o
n
)

80 77

Nomogram for determining the amount of ammonium sulfate which will yield the desired %
saturation of (NH
4
)
2
SO
4
. A saturated (NH
4
)
2
SO
4
solution is 4.1 M and 3.9 M at 25
o
and 0
o
,
respectively. Find the appropriate initial and desired final concentrations of (NH
4
)
2
SO
4
on
the vertical and horizontal scales, respectively. The point of intersection is the number of
grams needed per liter of solution. Multiply this number by the number of liters and add that
amount of (NH
4
)
2
SO
4
to the solution. (Modified from Methods of Enzymology 1:76).

Example: you have 100 ml of protein solution and would like to remove the proteins
precipitate at 35% saturation (NH
4
)
2
SO
4
and collect the remaining proteins precipitated at
60%. Add 20.9 grams of (NH
4
)
2
SO
4
to the 100 ml. Centrifuge at 10,000 x g for 10-20
minutes, discard the pellet and determine the volume of the supernatant. Multiply 164 by the
volume in liters and add that number of grams of (NH
4
)
2
SO
4
to the solution. Centrifuge as
above, discard the supernatant and dissolve the pellet in appropriate volume of buffer.


73
CHAPTER 9--CHROMATOGRAPHY

Grossly dissimilar molecules are relatively easy to separated. For example, lipids,
proteins and DNA can usually be separated from one another based on differences in solubility
in various solvents. Separation of substances with similar chemical and physical properties is
more complex and subtle. Although individual proteins are unique in terms of their structures,
the overall chemical and physical properties are somewhat similar in that they are all polymers
of amino acids. Therefore differential solubility has a limited ability to separate proteins.
Chromatography provides a means to refine the separation of substances.

BASIC PRINCIPALS

The basis of chromatography is to place substances to be separated into a system with
two phases: a mobile phase and a stationary phase. Substances are then separated based upon
their differential interaction with these two phases as the mobile phase moves across the
stationary phase. In the case of liquid chromatography the mobile phase is a solvent. Molecules
of interest (called the solute) are dissolved in the solvent and the solvent then flows across a
solid matrix (i.e., the stationary phase). Solutes interact with the stationary phase by reversibly
binding to the stationary phase. The strength of the binding between the solute and the
stationary phase will determine how fast the solute is carried by the mobile phase. For example,
substances which do not bind or interact with the solid phase will be carried unimpeded by the
solvent. Whereas substances that interact with the solid phase will be temporarily retained.
Therefore two substances that interact with the solid phase to different degrees can be separated
from one another.

In most applications the mobile phase is liquid. The major exception is gas-liquid
chromatography in which the mobile phase is a gas and the stationary phase is a liquid absorbed
to a solid support. In liquid chromatography, the stationary phase can be in a column
configuration or in a thin layer. The column is probably the most common way to hold the solid
support and is especially convenient for preparative work such as the isolation of the solute. In
particular, proteins are generally isolated by column chromatography. The column is a cylinder
or tube which holds the solid phase matrix and the liquid phase is passed through this column.

There are several distinct types of solid phases used in the isolation and analysis of
proteins (see Table). These various types of solid supports will separated proteins based upon
different chemical and physical properties (discussed below).

CHROMATOGRAPHY DISCRIMINATION
Ion Exchange Charge
Gel Filtration Size and Shape
Hydrophobic Surface Hydrophobicity
Reverse Phase Total Hydrophobicity
Affinity Specific Amino Acids
Adsorption Amino Groups?

74
EQUIPMENT

The equipment needed for column chromatography can range from expensive
workstations to pasteur pipettes. The central component is the column. A pump may be needed
to control the flow rate of buffers through the column. However, gravity can also be used. The
solvent used in the mobile phase will often need to gradually change. This gradual change is
accomplished by a gradient maker. The elution of substances can be monitored during
chromatography with an in line spectrophotometer which measures the absorbance of material
coming off the column. In the case of protein chromatography the detector/recorder monitors
the A
280
. Detectors to record radioactivity or fluorescence during chromatography are also
available.




If chromatography is used as part of a protein purification scheme a fraction collector
is needed. A fraction collector is a device that will automatically collect the liquid flowing from
the column in separate tubes. Most fraction collectors will allow fractions to be collected per
unit time or per unit volume (i.e., number of drops). In either case, the end result is a series of
tubes containing approximately equal volumes. The tubes can then be evaluated for the
substances of interest. The amounts of the substances being measured are often plotted against
either the fraction number, volume or time. These three variables can be easily inconverted from
the flow rate and volume of the individual fractions. Some fraction collectors can also be
interfaced with the detector/recorder and programmed to collect only the peaks.

High Performance (Pressure) Liquid Chromatography. HPLC is not a distinct
chromatographic technique, but an advancement of technology. The flow rates in
conventional chromatography is limited because of the compression of the support matrices
used in the columns. These low flow rates result in diffusion and loss of resolution. New
resins that can withstand packing and high flow rates allow for higher resolution have been
developed. This allows for separations to be carried out under higher pressures (i.e., high
flow rates) resulting in increased resolution. All of the same types of chromatographic media
75
are available for HPLC as in conventional chromatography. HPLC is more widely used for
the separation of small molecules but can be applied to the separation of proteins in some
applications.

Fast protein liquid chromatography (FPLC) is a similar concept as HPLC, but
specificially designed for protein separations. FPLC also uses special columns and pumps to
achieve high flow rates and therefore faster separations. In general the flow rates obtained
with FPLC are not as great as those achieved with HPLC.


ADSORPTION CHROMATOGRAPHY

The basic principal of chromatography discussed above refers specifically to adsorption
chromatography. Thin-layer chromatography (TLC) and paper chromatography are examples of
adsorption chromatography. These techniques are usually used to separate small molecules,
such as nucleotides, amino acids, lipids, simple carbohydrates, for analytical work. However, in
many applications HPLC in a column format is used. Adsorption chromatography is not widely
used in protein purification. Hydroxyapatite, a crystalline form of calcium phosphate, is an
example of a stationary phase medium that is sometimes used to analyze either proteins or
nucleic acids. The basis of protein separation on hydroxyapatite columns is not understood and
is difficult to predict based upon a protein's chemical and physical properties. It may involve
non-specific interations between the positive calcium and negative carboxyl groups on the
protein and the negative phosphate and positive amino groups on the protein. Chromatography
on hydroxyapatite columns has been used successfully in some cases to separate proteins that
were not readily separable by other techniques.

ION EXCHANGE CHROMATOGRAPHY

Ion exchange chromatography (IEC) is a specific type of adsorption chromatography
based upon charge-charge interactions. The stationary phase consists of fixed charges on a solid
support. The fixed charges on the stationary phase can be either negative or positive and are
respectively referred to as cation exchange or anion exchange chromatography. Counter ions
will interact with the fixed charged groups and can 'exchange' with solute molecules. In other
words, substances to be separated will replace the counter ions associated with the chroma-
tography medium and stably bind to the exchanger via electrostactic interactions. Conditions in
which some solute molecules are electrostatically bound to the exchanger and other solute
molecules are not bound can be used to separate solutes (Figure).

76

Basic Principal of IEC. A solution containing the solutes of interest (
+
YH & Z
+
) is
applied to the ion exchanger. The solutes bind to the exchanger and displace the
counter-ions (X
+
) already bound to the column. The solute is removed from the
stationary phase by increasing the concentration of counter-ions or by changing
the pH to affect the charge of the solute. Weakly charged solutes (
+
YH) will be
displaced from the stationary phase with lower concentrations of counter-ions than
more highly charged solutes (Z
+
). This results in separation of solutes based upon
their net charges

For example, mixtures of adenine nucleotides (i.e., adenosine, AMP, ADP and ATP)
can be separated by IEC. Adenosine, which is uncharged does not bind to the anion exchanger.
As an increasing concentration of formate is applied to the column, AMP is the first to elute
followed by ADP and then ATP. The order of elution corresponds to the overall negative charge
of the nucleotides (i.e., number of phosphate groups).



Proteins are complex ampholytes (have both - and +
charges). Negative charges are due to aspartic acid and glutamic
Amino a. pK
a

Asp, Glu 4.3-4.7
His 7
Lys, Arg > 10
77
acid residues and positive charges are due arginine, lysine and histidine residues. (See Appendix
of Introduction to Proteins for structure of side-chain groups.) Each of these side-chains
functions as either a weak acid or weak base and has a pK
a
value (see Table). The exact pK
a
for
the side-chain groups will depend on the position of the residue within the protein. The side-
chains of aspartate and glutamate have a carboxylic acid group that will either be negatively
charged or uncharged depending upon the pH. At pH values greater than the pK
a
the carboxylic
acid group will be deprotonated, or negatively charged, whereas below the pK
a
the carboxylic
acid will be protonated and thus have no charge. Similarly, the side-chains of the basic amino
acids have an amine group that will be positively charged (i.e., protonated) at pH values below
the pK
a
or uncharged (i.e. deprotonated) at pH values above the pK
a
.

The combination of all of the charged side chains will give the protein a net charge
which depends on the pH. The isoelectric point (Figure) of a protein is the pH at which the net
charge is zero (i.e., the number of positive and negative charges are equal). Therefore, proteins
will have either a net negative charge or net positive charge depending upon the isoelectric
point of the protein and the pH of the solution, and thus, it is possible to use either anion
exchange or cation exchange chromatography. Anion exchange chromatography is generally
carried out above the isoelectric point of the
protein of interest and cation exchange
chromatography is carried out below the
isoelectric point.

The functional groups in IEC can be
either strong or weak acids for cation
exchange or strong or weak bases for anion
exchange. A strong exchanger is used for the
separation of weakly ionizable groups and a
weak exchanger is usually used for more
highly charged solutes. In addition, weak exchangers result in better resolution when the charge
differences between solutes is small. IEC of proteins is most often carried out with weak
exchangers. The matrix should not have ionizable groups or bind proteins. Common matrices
used in IEC are dextran (trade name Sephadex), agarose (trade name Sepharose), and cellulose
are common matrices used in protein chromatography.

The basic steps in column chromatography are to prepare the column, apply the sample
to the column, and to elute the solute from the column. Prepared IEC columns can be purchased
or alternatively the ion exchange media is purchased and prepared. The amount of preparation
necessary will depend upon the form of the media and the manufacturer's instructions should be
followed. In general, short wide columns work best for IEC and that the total column volume
should be such that all of the protein binds in the top 1-2 cm of the column. The sample should
be loaded under conditions (i.e., appropriate pH and low ionic strength) which promote the
binding of the protein of interest. Proteins that do not bind to the column are washed away using
the same buffer.

Proteins are eluted by increasing the ionic strength or changing the pH. Increasing the
ionic strength can be done batchwise by sequentially washing with buffers containing higher
78
concentrations of salt or gradually using a gradient maker. Eluting in a 'steps' is generally
simpler, but of lower resolution. A gradient maker will mix two buffers of different ionic
strengths resulting in controlled
increase in the ionic strength of the
elution buffer. The slope and shape of
the gradient will determine the
resolution in regards to separating
solutes. For example, decreasing the
slope of the gradient will lead to a
greater separation of the solutes (see
Figure). However, as the slope
decreases the proteins will be more dilute.

Chromatofocusing is a specialized form of ion-exchange chromatography which
separates proteins according to their isoelectric points. This method takes advantage of the
buffering capacity of the exchanger and generates a pH gradient across the column. The column
is developed with a mixture of buffers called the 'polybuffer'. As these migrate down the
column, the most acidic components will bind to the basic groups of the exchanger and lead to
an increase in the local [H
+
]. As elution progresses the pH as each point in the column is
gradually lowered due to the addition of more buffer components. The more acidic components
will elute, migrate further down the column and bind to again to the basic exchanger groups.
This will result in a pH gradient across the length of the column. The sample is loaded onto the
column at a pH > than the pI of the solute of interest. As the pH drops the protein elutes and
migrates down the column until the pH is greater than the isoelectric point and then reabsorbs.
This will continue until the proteins elute at their isoelectric points.

79
HYDROPHOBIC CHROMATOGRAPHY

Proteins can be separated according to
differences in their hydrophobicities. Hydropho-
bicity is a chemical property which promotes the
aggregation of nonpolar compounds with each
other in an aqueous environment. These hydro-
phobic interactions are not an attractive force per
se, but are forced upon nonpolar compounds in a
polar environment. The media for hydrophobic chromatography is a support matrix such as
agarose with long chain hydrocarbons covalently bound. Two common examples are octyl-
agarose (8 contiguous methyl groups) and phenyl-agarose (Figure). This will provide a hydro-
phobic surface for proteins to interact with instead of aggregating with each other. Highly
hydrophobic resins like octyl-agarose are best for weakly hydrophobic proteins, whereas less
hydrophobic resins like phenyl-agarose are better for proteins of intermediate hydrophobicity.

Proteins bind to hydrophobic columns under conditions that promote hydrophobic
interactions and these conditions will determine the extent of binding. For example, raising the
ionic strength increases hydrophobic interactions, or the 'salting out' effect (see chapter on
Differential Solubility). Both anions and cations can be listed in a series in terms of either
promoting hydrophobic interactions or increasing the chaotropic effect. Chaotropic agents
disrupt the structure of H
2
O by decreasing H-bonding and therefore decrease hydrophobic
interactions. The most common salt used in hydrophobic chromatography is (NH
4
)
2
SO
4
. The
(NH
4
)
2
SO
4
concentration should not promote protein precipitation (eg., 20% saturated or
approximately 1 M). Other salts can be substituted, but higher concentrations may be required
(eg., 4 M NaCl).

Conditions which decrease hydrophobic interactions
are used to elute proteins from hydrophobic columns (Box).
Decreasing the ionic strength is generally the preferred
method since the other methods introduce substances which
may be difficult to remove and/or denature the protein.
Decreasing the ionic strength can be done in a step-wise fashion or with gradients as discussed
for ion exchange chromatography. Substances which affect the polarity of the solvent (i.e.,
water) will also affect hydrophobic interactions. For example, a gradient of ethylene glycol will
lead to the differential elution of proteins. Similarly, chaotropic agents can be use to elute
proteins from hydrophobic columns. Detergents are more hydrophobic than proteins and will
complete with the proteins for binding to the hydrophobic matrix.

decrease ionic strength
decrease solvent polarity
chaotropic agents
detergents
increasing salting-out effect
anions: PO
4
, SO
4
, Cl, Br, NO
3
, ClO
4
, I, SCN
cations: NH
4
, Rb, K, Na, Li, Mg, Ca, B
increasing chaotropic effect
80

Reverse phase chromatography (RPC)
also separates compounds based upon differences
in hydrophobicities. RPC differs from hydrophobic
interaction chromatography (HIC) in that the
mobile phase is a nonpolar solvent such as hexane
instead of an aqueous salt solution. HIC is usually
performed under non-denaturing conditions and
separates proteins according to differences in sur-
face hydrophobicity. RPC is carried out under denaturing conditions and separates according to
differences in the total hydrophobicity, since all of the amino acid residues are available for
interaction with the stationary phase. The separation of small polypeptides and proteolytic
fragments is a common application of RPC.

GEL FILTRATION CHROMATOGRAPHY

Gel filtration, also called molecular sieve
chromatography or size exclusion chromatography,
separates proteins on the basis of molecular size. A
protein solution is passed over a column made up of
small beads composed of cross-linked polymers.
The degree of cross-linking will defined a pore size.
Solutes larger than this pore size are excluded from
the matrix and pass through the column unimpeded.
In other words they do not enter the beads but flow
around them. Smaller solutes will enter the gel matrix and are retained on the column longer.
The retention time is inversely proportional to the size of the solute.

Several different polymers have been used as support
matrices for gel filtration chromatography (Box). Different
grades of gel filtration media with different pore sizes are avail-
able. The different pore sizes allow for separation of macromole-
cules with different size ranges. Gel filtration chromatography is generally carried out in buffers
containing 0.15-1.0 M salt to prevent interactions of proteins with the support matrix. Unlike
the other chromatographic methods the solute (eg., protein) does not bind to the stationary
phase during chromatography. Therefore, gel filtration is a gentle technique and fragile proteins
are not damaged by adsorption to the chromatographic support. Since proteins do not bind to
the stationary phase, the resolution is dependent upon loading the smallest possible volume of
sample.

HIC vs. RPC

Hydrophobic
Reverse
Phase
Mobile
Phase
Polar Solvent
Nonpolar
Solvent
Conditions Native Denatured
Solute
Properties
Surface
Residues
Total
Residues
Dextran (=Sephadex)
Agarose (=Sepharose)
Polyacrylamide
81
Gel filtration can also be used to determine the molecular
weight of a protein if the columns are calibrated by using
molecular weight standards. Proteins of known size are passed
over the column and the K
av
is determined for each protein
according to the following equation:

K
av
= V
e
- V
o
/V
t
- V
o
.

The void volume (V
o
), also called the excluded
volume, is the elution volume of a substance which
is two large to enter the matrix of the support
medium. This is experimentally determined with a
standand known to be completely excluded and
represents the solvent between the beads. The total
volume is calculated from the volume of the
column bed (r
2
x length) and represents the elution
volume of molecules completely included in the
matrix. All substances, assuming there are no
interactions with the matrix, will elute in this
volume. The K
av
values are plotted against the log
of the molecular weight for each protein standard. The molecular weight of the unknown
protein can then be determined from its K
av
and the standard curve.

Shape also affects the retention of macromolecules. For example, long rod shaped
proteins elute from gel filtration columns at apparent molecular weights greater than their actual
molecular weights.

Desalting. Gel filtration is often employed to remove salts or other small molecular
weight solutes from protein solutions or to carry out buffer exchanges. The sample is loaded
onto a gel filtration column that will exclude proteins (eg., G-10 or G-25) and the void volume
collected. Desalting columns are much faster than dialysis. However, the sample is usually
diluted 2-4 fold. Spin columns, widely used in DNA isolations, are another example of desalting
columns.

AFFINITY CHROMATOGRAPHY

The biological function of proteins often involves binding to
or interacting with specific ligands, substrates, co-factors, inhibitors,
other proteins, etc. Affinity chromatography takes advantage of these
specific interactions between proteins and ligands. A protein mixture
is passed over the column with an appropriate moiety covalently
attached to a solid support. Only the protein which recognizes the
ligand of interest will bind to the column and the other proteins will
be washed away. Affinity chromatography can be very specific and
may allow the isolation of a protein in a single step.

V
o
= void volume
V
t
= total volume
V
e
= elution volume
82
Affinity columns are usually prepared from cyanogen bromide (CNBr)-activated
agarose. The CNBr group reacts with primary amines and covalently attaches the ligand to the
solid support. In some cases, the ligand already bound to the matrix is commercially available.
In addition, linker arms are often used to extend the ligand away from the matrix of the solid
support. The matrix should not adsorb contaminating proteins and covalent attachment of ligand
to matrix should not alter its binding to protein of interest. Binding of the protein to the ligand
should be relatively tight (i.e., high affinity), but at the same time the binding should not
preclude the ability to elute the solute from the column.

Conditions which promote the dissociation of the protein and
ligand are used to elute proteins from affinity columns (box). For
example, high concentrations of the free ligand can compete with the
bound ligand resulting in elution. In cases where affinity elution is not
possible or does not work, it may be possible to change the pH or
increase the ionic strength resulting in a destabilization of the protein-ligand interactions.
Another possibility is to use denaturing or chaotropic agents to elute the protein which will
result in the protein unfolding. However, these later methods can only be used if it is not
important to recover protein in the native state.

http://www.science.uts.edu.au/subjects/91326/Section3/section3.html (good summary of
chromatography)

affinity elution
pH changes
ionic strength
chaotropic salts
denaturing agents
83
CHAPTER 10--MEMBRANES AND DETERGENTS

Membranes are composed of a lipid bilayer and associated proteins. Lipids are
amphipathic molecules in that they contain a polar head group and hydrophobic tails. In
aqueous solutions lipids will aggregate such that the hydrophobic tails interact with other
hydrophobic tails and the polar head groups are exposed to water. The two possible
configurations are: 1) a spherical micelles with the hydrophobic tails pointed inward, or 2) a
bilayer with the hydrophobic tails sandwiched between the polar head groups. The shape of the
lipid and its amphipathic nature cause them to spontaneously form bilayers in aqueous solutions
and accounts for the stability of membranes.

Proteins interact with this layer bilayer in several different fashions. Transmembrane
proteins pass through the bilayer. A hydrophobic domain, typically an -helix composed of
amino acids with hydrophobic side-chains, interacts with the hydrophobic tails of the lipids and
anchors the protein to the bilayer. Some transmembrane proteins will have multiple membrane
spanning domains. Other membrane proteins are anchored to the lipid bilayer via fatty acids or
lipids that are covalently attached to the protein. Other membrane are attached to the membrane
by non-covalent association with other membrane proteins.

Many cellular processes occur on membranes or in membrane-bound subcellular
compartments. Approximately half of a cell's total protein is associated with membranes or
found in membrane-bound compartments. The study of these membrane associated processes
may require the isolation of membranes or membrane proteins. The choice of tissue and
membrane fraction (i.e., organelles) will depend in part on the phenomenon being studied.
Some cell types are better sources for certain types of membranes. The cells need to be
disrupted by procedures which preserved the activity
of interest (see Appendix). Disruption of cells or
subcellular compartments will usually result in the
membranes forming vesicles. The different types of
membranes will have different sizes and densities
based upon their lipid and protein composition and
can be prepared by differential centrifugation, density
gradient centrifugation or a combination of the two.
choice of cells or tissue
choice of membrane fraction
homogenization conditions
preparation of membranes
differential centrifugation
density gradient centrifugation
solubilization of membranes
isolation of proteins
84
The exact isolation technique will depend upon the source and type of membrane being isolated
as well as the desired purity.

TYPES OF MEMBRANE PROTEINS

It is possible to remove proteins from the lipid bilayer and subject membrane proteins to
further analysis. The method for solubilization will depend on the nature of the proteins
association with the lipid bilayer. In this regard, membrane proteins can be viewed as either
integral or peripheral. Integral membrane proteins include the transmembrane proteins and
lipid anchored proteins. Such proteins are solubilized under conditions which disrupt the lipid
bilayer. Peripheral membrane proteins are non-covalently associated with other membrane
proteins and can be differentially extracted under conditions which disrupt protein-protein
interactions, but leave the bilayer intact (eg. high salt, low salt, alkaline pH, urea).

Differential Solubility of Erythrocyte Membrane Proteins

Erythrocyte membranes were suspended in 0.1 mM EDTA (pH 9.2) and centrifuged. The pellet was
resuspended in 0.1 mM EDTA (pH 11) and centrifuged again. The total extract (t), the first supernatant
(s), and the first (p1) and second (p2) pellets were examined by SDS gel electrophoresis (see chapter on
Electrophoresis). Integral membrane proteins such as band 3 and glycophorins are not extracted from
the lipid bilayer under either condition. Ankyrin and band 4.1, which bind directly to band 3 and
glycophorin respectively, are extracted at pH 11 but not pH 9.2. Spectrin and actin, which form a two-
dimensional network connected to the membrane via interactions with ankyrin and band 4.1, are
extracted at pH 9.2.

Membrane proteins can be selectively extracted from the membrane depending on the
nature of their association with the lipid bilayer. For example, some peripheral membrane
proteins can be extracted from the membrane by treating with high salt (0.5-1.0 M) solutions.
This will remove proteins that interact weakly with the bilayer via electrostatic interactions. The
integral membrane proteins and other peripheral membrane proteins remain associated with the
lipid bilayer and are removed by centrifugation. Most peripheral proteins will require low ionic
strength and alkali pH (< 9) for solubilization. The exact pH will depend on the particular
protein (see figure for example). Chaotropic agents, such as urea, can also be used to solubilize
peripheral membrane proteins. However, such agents will not exhibit selectivity and many will
disrupt the bilayer and extract integral membrane proteins. Integral membrane proteins can be
selectively extracted through the use of detergents (see below).
85

The optimal conditions for solubilization will need to be determined for each individual
protein. Isolated membranes are mixed with the extraction buffer. The temperature and time can
also be varied. Following the incubation, the particulate (i.e., pellet) and soluble (i.e,
supernatant) fractions are separated by centrifugation and the two fractions analyzed for the
protein of interest. In general, extraction with detergents will disrupt the bilayer resulting in a
particulate fraction that represents cytoskeletal elements and associated proteins. However,
some detergents, especially ionic detergents, will also disrupt protein-protein interactions
involved in formation of cytoskeletons. Extraction with low ionic strength buffers at alkaline pH
will result in a pellet consisting of the bilayer still containing the integral membrane proteins.
Sequential extractions can be used to remove proteins that are not of interest before solubilizing
the protein(s) of interest.

The extraction conditions may preclude the analysis of enzyme or other functional
activities. It may be possible to restore the functional activity by restoring the samples to
appropriate conditions or removing the detergents.

DETERGENTS

Detergents are amphiphilic compounds, meaning that they contain both hydrophilic and
hydrophobic moieties. A wide variety of chemicals can be classified as detergents. These can be
subgrouped based upon the natures of their hydrophobic and hydrophilic regions. Two types of
hydrophobic regions found in biologically important detergents are alkyl chains and steroid
skeletons (eg., deoxycholate). The alkyl chains can be straight or branched and some also
contain phenyl (eg., NP-40) or cyclohexyl groups. The polar groups can be classified as anionic,
cationic, amphoteric (i.e., zwitterionic), and non-ionic. Common non-ionic polar groups are
glucosides and polyoxyethyl groups. In general, the non-ionic detergents are less denaturing
than the ionic detergents.

Detergents have surface activity and readily
form micelles that are soluble in H
2
0 (unlike
lipids). Micelles are aggregates of detergent
molecules in which the polar groups are exposed to
the aqueous environment and the hydrophobic
groups form an inner core.

Terms defining the properties of detergents are: critical micellar concentration, micellar
molecular weight, critical micellar temperature, and cloud point. The critical micellar
concentration (CMC) is defined as the minimum concentration at which the detergent forms
micelles. The micellar molecular weight is the average micelle size of a pure detergent and
represents the number of detergent monomers within a micelle. The micellar M
r
usually
expressed as an aggregation number (N) which is the average number of monomers in one
micelle. In general, detergents with a low CMC have a high micellar M
r
. The CMC and micellar
M
r
are affected by temperature, pH and ionic strength. Increasing the ionic strength tends to
lower the CMC and raise the micelle size. The critical micellar temperature is the minimum
temperature at which a detergent will form micelles. The cloud point is the temperature above
86
which detergent micelles will form super aggregates. Detergents should be used under
conditions which are optimal for micelle formation since micelles are critical for detergent
function.

Examples of Detergents and Their Properties
Name Structure CMC N (MW)
Nonidet

(N) P-40

0.3 mM
100-155
(647)
sodium dodecyl
sulfate (SDS)
8.3 mM 62 (288)
sulfobetaine
(SB12)
3.6 mM 55 (336)
n-octylglucoside

14.5
mM
20-25
(292)
deoxycholate

20 mM
3-12
(417)
A few examples of detergents illustrating the different types of hydrophillic and hydrophobic groups.
Critical micellar concentrations (CMC) and aggregation numbers (N) are also shown. MW is the
average molecular weight of the monomer in daltons.

Detergents interact with both membrane lipids and proteins. At low detergent to lipid
ratios, detergent monomers incorporate into the lipid bilayer without disrupting the membrane.
As the detergent:lipid ratio increases, lipids, because of their similar properties as detergents,
will form mixed micelles with the detergent. Therefore, the detergent concentration needs to be
above the CMC. Detergent molecules also interact with the hydrophobic portions of the proteins
and in effect replace the lipids. This will lead to the proteins being coated with detergents and
prevent protein-protein interactions that would normally result in protein precipitation. Protein
solubilization occurs at or near the CMC for most detergents.
There are no general rules for choosing among the
various types of detergents. It is not possible to predict which
detergent will be most useful for any particular application.
Pilot experiments to determine the optimal detergent as well as
the optimal conditions for maximal protein solubilization are
carried out. Membrane samples are incubated with various
Possible Detergent Effects
protein structure
protein activity
interference with assays
separation techniques
87
concentrations of detergents for appropriate times and detergent insoluble material is removed
by centrifugation. The soluble and particulate fractions are analyzed for the protein of interest.
Ideally, detergents should not affect the native structures or activities of the protein(s) being
characterized. In addition the detergent should not affect the measurement of that protein (eg.,
assay) or chromatography systems that will be utilized in the isolation
of that protein.

DETERGENT REMOVAL

It is often necessary to use a detergent to solubilize the protein of interest. However,
later it may be necessary to remove the detergent. In general, detergents are difficult to remove.
Therefore, the ease at which a detergent can be removed is also an important criteria to consider
when choosing a detergent.

Various methods for detergent removal have been described (Box). The choice of
method will depend somewhat on the class of detergent being removed. Dialysis or desalting
columns can be used for ionic detergents with a small micelle size and high CMC. However,
detergents with a low CMC tend to have a large micelles size and will not pass through the
dialysis membranes. Dialysis would require substantial dilution for the removal of the detergent
monomers. Gel filtration can be used to remove the large micelles from the smaller proteins.
The proteins will still be coated with detergent though.

Hydrophobic chromatography can be also used to remove detergent. Similarly, in the
case of ionic detergents, urea can be added and ion exchange chromatography can be carried
out. In the presence of urea proteins will not bind to the column. Conversely, ion exchange
chromatography or affinity chromatography can be used to bind proteins of interest. The
column is washed extensively and the proteins are eluted in detergent free buffers.

In many cases complete removal of the detergent will result in protein precipitation and
is undesirable. Detergent replacement can be used in situations where a detergent which is
optimal for solubilization interferes with the analysis or further purification of the protein. The
sample is subjected to dialysis or chromatography in the presence of the second detergent.
Another possibility is to replace the detergent with urea.
Dialysis
Chromatography
Replacement
88
APPENDIX 1. CELL DISRUPTION TECHNIQUES

Organisms can be studied on many
levels. The whole organism can be studied
(eg., behavior, toxicity, etc.). Tissues or
organs can be removed by dissection. Tissues
can be broken down to single cells by grinding
or enzymatic digestions. In addition, single
cells can be obtained through in vitro cell
culture or from single-celled organisms. The
choice of organism, tissue, or cell type will
depend on the research topic or the protein of
interest. A particular organism, tissue or cell
type is often the subject of the research. In
other cases, though, a particular phenomenon
is the object of research. Therefore, the choice
of organism, tissue, etc. will depend upon the biological phenomenon being studied. Different
tissues and cells will express different amounts of macromolecules or organelles. For example,
liver is good source of mitochondria and erythrocytes are a good source of purified plasma
membrane.

To study biology at the subcellular and molecular levels it is necessary to disrupt or lyse
the cells. Many different methods can be used to disrupt cells (Table). The various techniques
all have advantages and disadvantages and the method used will depend on the type of cell and
the application. Cell disruption/lysis may destroy the organelle or denature the protein of
interest. Therefore, it is important to have an assay to evaluate the disruption techniques in
terms of preserving the structures or activities of interest. In general, one needs to empirically
determine the optimal conditions for cell disruption.

Subcellular organelles are isolated after cell lysis by differential and/or density gradient
centrifugation. Generally homogenation, presses or nitrogen cavitation are the appropriate
methods for cell lysis if intact organelles need to be isolated. The other methods are most
appropriate in cases were macromolecules are being evaluated without prior isolation of
organelles.

89

Cellular Disruption Methods
TECHNIQUE COMMENTS
homogenation
Cells are placed in homogenizer or blender and disrupted by shear forces or
grinding. Many different types of homogenizers are available depending on the
application. Most consist of a pestle with a defined clearance and can be hand
or motor driven. Homogenation can be combined with osmotic methods.
Heating can be a problem.
presses

Cells are placed under high pressure in a stainless steel chamber. After
reaching a defined pressure the chamber is opened and the cells exit through
a small orifice at a controlled rate. The rapid exposure to atmospheric
pressure causes the cells to lyse. Optimal pressure for cell disruption needs to
be determined empirically for each type of cell and experimental condition.
N
2
-cavitation

Cells are placed under high pressure in a N
2
atmosphere. The pressure is
suddenly released, causing the N
2
dissolved within cells to boil off and rupture
the cells. Both the pressure and rate of pressure release can be controlled.
osmotic,
hypotonic

Cells are placed in an isomolar solution of a permeable solute or a hypotonic
solution. Water will enter cell which results in swelling until the membrane
ruptures. The technique is gentle, but may not lyse cells with walls or rigid
cytoskeletons (eg., bacteria, plants, etc.).
sonication
Ultrasonic waves are used to disrupt cells. Heating is a problem and the
process tends to vesiculate membranes and subcellular compartments.
freeze-thaw
Ice-crystals formed during freezing will rupture cell membranes. It will
generally take multiple cycles of freeze-thaw to disrupt all of the cells.
detergents

Detergents will disrupt the lipid bilayer allowing the contents to be released
and will solubilize membrane proteins. Many detergents denature proteins and
they are often difficult to remove.
chaotropic
agents
Chaotropic agents solubilize proteins by disrupting the structure of water and
minimizing hydrophobic interactions. Protein denaturation is a problem.
enzymatic
Bacteria, yeasts and plant cells are often treated with enzymes to remove the
cell wall. Enzymatic treatment is often combined with other disruption
methods.

91
CHAPTER 11--ELECTROPHORESIS

Electrophoresis, like centrifugation, is a hydrodynamic technique. A charged particle
(i.e., molecule) in an electric field experiences a force that is proportional to the potential
difference (E), or voltage, of the electric field and inversely proportional to the distance (d)
between the electrodes. (The potential difference divided by the distance (E/d) is referred to as
the field strength.) The force will also be proportional to the net charge of the molecule (q).
Therefore, the force experienced by the molecule can be expressed by the following equation:

F = Eq/d

This force will by opposed by a
frictional force (= fv) where f is a frictional
coefficient and v is the velocity of the particle.
The frictional coefficient depends on the size
(eg., r = radius) and shape of the molecule and
the viscosity () of the medium. For example, in the case of a sphere the frictional force is:

F
f
= 6rv

A particle will move at a velocity (v) so that these two forces are equal, therefore:

6rv = Eq/d or solving for v

v = Eq/d6r

This equation indicates that the mobility (i.e., velocity) of a molecule in an electric field
is proportional to the electric field (E/d), or more simply the applied voltage, and the net charge
of the molecule. The mobility is inversely proportional to a frictional coefficient (i.e., size and
shape of the molecule and the viscosity of the medium), as indicated by the following equation:

mobility = (applied voltage)(net charge)/(friction coefficient)

Therefore, it is possible to derive information about the charge, size and shape of a molecule by
its mobility in an electric field.

GEL ELECTROPHORESIS

Electrophoresis of macromolecules can be carried out in solution. However, the ability
to separate molecules is compromised by their diffusion. Greater resolution is achieved if
electrophoresis is carried out on semi-solid supports such as polyacrylamide or agarose gels.
Gels are formed by cross-linking polymers in aqueous medium. This will form a 3-dimensional
meshwork which the molecules must pass through. Polyacrylamide is a common gel for protein
electrophoresis whereas agarose is more commonly used for nucleic acids (see Chapter 18).
Agarose gels have a larger pore size than acrylamide gels and are better suited for larger
macromolecules. However, either type of gel can be applied to either nucleic acids or proteins
92
depending on the application.

Gels are formed from long polymers in a cross-linked lattice (Figure). The space
between the polymers are the pores. Higher concentrations of the polymer will result in smaller
average pore sizes. Polyacrylamide gels are formed by covalently cross-linking acrylamide
monomers with bis-acrylamide with a free radical like persulfate (SO
4
). The cross-linking of
the acrylamide polymers results in 'pores' of a defined size. The total acrylamide concentration
and the ratio of bis-acrylamide to acrylamide will determine the average pore size. The
polyacrylamide solution is poured into a mold and polymerized. This mold can be a cylindrical
tube, but is usually a 'slab' poured between two glass plates (see below).


Diagramatic representation of gels. Gels are form by cross-linking polymers into a 3-dimensional mesh-
work or lattice (left). This results in a pourous semisolid material that solutions can pass through. In other
words the open spaces are filled with solvent. Increasing the concentration of the polymer will resulting in
smaller 'pores'. Agarose gels are formed by heating the desired concentration of agarose in an aqueous
buffer. After the agarose dissolves, the solution is cooled and the gel forms. Polyacrylamide gels are
formed by chemically crossing acrylamide and bis-acrylamide using a free-radical such as persulfate
(right).

Since the gel is solid with respect to the mold, all molecules are forced through the gel.
Smaller molecules will be able to pass through this lattice more easily resulting in larger mole-
cules having a lower mobility than smaller molecules. In other words, the gel acts like a mole-
cular sieve and retains the larger molecules while letting the smaller ones pass through. (This is
opposite of gel filtration where the larger molecules have a higher mobility because they to not
enter the gel.) Therefore, the frictional coefficient is related to how easily a protein passes
through the pores of the gel and size will be the major determinant of the mobility of molecules
in a gel matrix. Protein shape and other factors will still affect mobility, but to a lesser extent.
Substituting size for the frictional coefficient results in:

mobility (voltage)(charge)/(size)

In other words, the mobility of a protein during gel electrophoresis is primarily a function of its
charge/mass ratio.
93
EQUIPMENT

Equipment to conduct gel electrophoresis is relatively simple. They consist of a mold to
form the gels, an apparatus to hold the gel and contain buffers, and a power supply capable of
delivering the required voltage or current. There are many types of apparati for carrying out
electrophoresis depending on the application. Gels can be either in a vertical or horizontal
configuration. Polyacrylamide gels are run in a vertical fashion and agarose gels tend to be run
in a horizontal position.

Gels can either be formed as cylinders by using glass tubing as a mold (often called tube
gels) or formed as rectangular slabs. These slab gels are formed by polymerizing the acryamide
solution between glass plates separated by spacers (Figure). Typically the gel is 0.75-1.5 mm
thick. At the top a 'comb' is used to form sample wells. Slab gels allow multiple samples to be
compared on the same gel, thus eliminating gel-to-gel variations.




The formed gel is placed into the apparatus to that the top and bottom of the gel are in
contact with chambers containing buffer (Figure). These chambers contain electrodes which are
connected to a power supply. Thus an electric field is generated across the gel when a voltage is
applied. The buffer in the chambers is generally different that the buffer making up the gel for
protein electrophoresis and in some applications the buffers in the lower and upper chambers
may be different. In most applications the buffers are such that the protein has a negative charge
and therefore the anode (positve pole) will be in the lower chamber and the cathode (negative
pole) will be in the upper chamber. However, there are applications in which the proteins of
interest may be positively charged and therefore the electrodes will be reversed.

Discontinuous or "disc" electrophoresis. The Laemmli discontinuous buffers are exten-
sively used in gel electrophoresis. Discontinuous gels consist of two distinct gel regions referred
to as stacking gel and separating gel (see Table) and a Tris-glycine tank buffer. The stacking gel
94
has a lower acrylamide concentration, a lower pH and a lower ionic strength than the separating
gel.


The lower ionic strength of the
stacking gel results in a greater local elec-
tric field strength than in the separating gel.
The field strength difference combined with the lower acrylamide concentration results in
proteins having a higher mobility in the stacking gel than in the separating gel. In addition, the
glycine in the tank buffer has a higher mobility in the separating gel than in the stacking gel
because of the pH differences. Therefore, proteins will migrate faster than the glycine in the
stacking gel. When proteins reach the separating gel their mobility is decreased because of the
increased acrylamide concentration and decreased field strength, whereas the increase in pH
results in glycine having a higher mobility. All of these factors result in the proteins becoming
compressed at the interface between the two gels and thus increasing resolution (Figure).
Resolution in non-discontinuous electrophoresis depends partially on the volume of the sample.
However, stacking also occurs at the interface of the sample and gel, especially if a high voltage
is applied.

SDS-PAGE

Polyacrylamide gel electrophoresis in the presence of SDS (sodium dodecyl sulfate) is
the most common form of protein gel electrophoresis. SDS completely disrupts protein-protein
interactions and denatures almost all proteins resulting in a complete unfolding of proteins. In
addition, -mercaptoethanol (or other reducing agents) is often used to break disulfide bonds.
The SDS binds to the unfolded proteins giving all proteins a similar shape (i.e., random coil or
extend conformation) and an uniform charge-to-mass ratio. In other words, coating proteins
with a negatively charged detergent minimizes the effects of a protein's net charge. Therefore,
during electrophoresis in the presence of SDS the mobility of a protein now depends primarily
upon its size (i.e., mobility is inversely proportional to protein mass).

Mobility in SDS gel electrophoresis is expressed as a relative mobility (R
f
). The
distance the protein migrated is compared to the length of the gel, or:

R
f
= distance protein migrated gel length

The length of the gel is often defined by the migration of a substance which is not impeded by
the matrix such a small molecular weight tracking dye (eg., bromophenol blue). This mobility
can then be used to calculate the size of proteins. Protein standards of known size are used to
Composition of Laemmli Gels
Stacking
Gel
Separating
Gel
Acrylamide 3-4.5% 6-20%
pH 6.8 8.8
Ionic
Strength
0.125 M
Tris
0.375 M
Tris
95
generate a standard curve by plotting the log of the
molecular weight against the R
f
values. The molecular
weight of an unknown protein can be extrapolated from
its R
f
value (see Appendix 1). Such a calculated
molecular weight is designated as M
r
to indicate that it
is a relative molecular weight based on comparisons to
other proteins. For some proteins, though, this estimated
molecular weight can differ from the actual molecular
weight. In particular, highly charged proteins behave
anomalously during SDS gel electrophoresis. In
addition, some proteins are not completely denatured by
SDS and this retention of some structure will lower the
mobility.

Practical considerations. The first step in
electrophoresis is to pour the separating gel (Box). Pre-
poured gels are also commercially available. Separating
gels will typically contain 6-20% acrylamide. The size
range of the proteins being separated, the desired
resolution and the amount of sample being applied are
factors to consider when choosing an acrylamide con-
centration. Gradient gels can be used in situations
where it is necessary to examine both high and low
molecular weight proteins on the same gel. The stack-
ing gel is poured after the separating gel polymerizes
and just before electrophoresis to minimize diffusion
between the two gels.

Proteins to be analyzed by SDS-PAGE are solubilized in a
sample buffer that typically contains 2% SDS and 5% -mercapto-
ethanol and then boiled. The reducing agent is omitted in situations
where disulfide bonds need to be preserved. In situations where an
enzyme activity will be measure following electrophoresis (see
below) a lower SDS concentration is used and the sample is not boiled. The amount of protein
that can be loaded onto a gel is limited. Overloading the gels results in the pores becoming
plugged and has an adverse effect on the electrophoresis.

After loading the samples into the wells of the gel an electric field is applied across the
gel. The mobility of proteins in an electric field is proportional to field strength (E/d), or simply
the voltage (E) since the distance (d) is determined by the electrophoresis apparatus. Electro-
phoresis will proceed faster, and therefore finish sooner, at higher voltages. However,
electrophoresis generates heat in proportional to the amount of power, which is the product of
voltage and current (P = EI). Excessive heating may result in proteins precipitating within the
gel or have other deleterious effects on proteins depending on the nature of the protein sample
and total protein concentration. These electrical parameters change during electrophoresis
because the ions migrate to the anode and cathode buffers, and therefore lead to an increase in
1. Pour separating gel.
2. Pour stacking gel.
3. Load samples.
4. Apply electric field.
5. Stain or process gel.
96
resistance (R). Resistance affects the voltage and current (E = IR) depending upon which
variable is held constant. Most power supplies are capable of delivering constant voltage (E),
constant current (I), or constant power (P). Electrophoresis is usually carried out under constant
voltage or constant power to minimize the resulting increase in heating that occurs during
electrophoresis.

A tracking dye (bromophenol blue) is included in the sample. When this dye reaches the
bottom of the gel or some predetermined time afterwards the power is turned off and the
proteins detected. A common way to detect proteins after electrophoresis is to stain the gel with
Coomassie blue, a dye that binds proteins. Gels are usually 'fixed' before staining with an acetic
acid and methanol solution which precipitates proteins into the acrylamide matrix.

ISOELECTRIC FOCUSING

Isoelectric focusing (IEF) separates proteins
based on their isoelectric points. The isoelectric
point is defined as the pH at which a protein has no
net charge (i.e., the number of negative and positive
charges are equal) and is a measure of the protein's
net charge. Separating proteins according to their net
charge is accomplished by generating a pH gradient
in an electric field. The effect of protein size on
mobility is minimized by carrying out the electro-
phoresis gels with large pore sizes such as low
acrylamide concentrations (eg., 3.5%) or agarose.
This large pore size minimizes the molecular
sieving.

A pH gradient is generated with carrier
ampholytes. These ampholytes are a mixture of
aliphatic amines and either carboxylic or sulfonic
acid. They have a high buffering capacity, low
molecular weight (300-600 Da) and a range of pK
a
values. Initially the pH of an ampholyte
solution will be the average of the pK
a
values of the mixture. Application of an electric current
will cause the ampholytes to migrate toward the electrodes according to their charges.
Ampholytes that have pK
a
values above the pH will be positively charged and those with pK
a

values below the pH will be negatively charged. As the ampholytes migrate this will result in
changes in the local pH due to the buffering action of the ampholytes. This change in the local
pH will affect the charge on the ampholytes depending upon the pK
a
. The ampholytes will
continue to migrate until they reach a position in which the local pH equals their pK
a
(i.e., no
net charge). The end result is a pH gradient in which the most basic ampholytes are found at the
cathode, a dilute alkali solution (eg., NaOH), and the most acidic ampholytes are at the anode, a
dilute acid solution (eg., H
3
PO
4
). Carrier ampholytes with defined pH ranges can be purchased
or prepared by isoelectric focusing.

Proteins are also ampholytes and will migrate within the pH gradient until they reach a
97
pH equal to their isoelectric point. The carrier ampholytes are needed since the protein concen-
tration is generally not high enough to establish a stable pH gradient and the isoelectric points of
the proteins may not be uniformly distributed along a pH gradient.

Sample preparation is important for IEF in that many rea-
gents can adversely affect isoelectric focusing. In particular, the
ionic strength should be as low as possible. Precipitation of pro-
teins with acetone is a method for removal of excess salts, as well
as concentrating the protein. Separations can be performed under
either native or denaturing conditions. Urea is the preferred denaturing agent since it is
uncharged. Similarly, non-ionic detergents, such as Triton X-100 or NP-40, are less likely to
interfere with the formation of pH gradients. Protein precipitation is sometimes a problem in
that proteins tend to be less soluble at their isoelectric points and their local concentrations can
be quite high. In addition, the high voltages and high resistance (see below) associated with IEF
generates substantial heat which increases protein precipitation. Many apparatuses for IEF will
have a cooling mechanism to disperse the excess heat. Inclusion of urea and NP-40 in the gels
will also minimize protein precipitation.

IEF is an equilibrium phenomenon since the components of the system migrate until
they have no net charge. As the system approaches equilibrium the resistance approaches
infinity since there are no ions to conduct the current. However, the pH gradient will start to
break down before true equilibrium is reached and the ampholytes will migrate into the anode
and cathode buffers. This gradient breakdown is accompanied by a lowering of the resistance.
Therefore, the progress of IEF can be followed by performing the electrophoresis under
constant voltage and monitoring the current. Initially the current will rapidly drop in concor-
dance with the rapid migration of the ampholytes. As the ampholytes lose their net charge, the
resistance increases and the current decreases (E = IR). The rate at which the current decreases
levels off as the system approaches equilibrium. The current will start to rise again when the pH
gradient starts to break down. IEF needs to be discontinued before this point.

The pH gradient can be determined with marker proteins with known isoelectric points
or by measuring the pH along the gel. This is accomplished by slicing the gel into pieces,
eluting the ampholytes into distilled water and measuring the pH.

TWO-DIMENSIONAL GEL ELECTROPHORESIS

Conventional electrophoresis separates proteins according to their charge/mass ratios.
SDS-PAGE separates proteins according to subunit, or polypeptide, mass. IEF separates
proteins according to isoelectric point (or charge). It is possible to sequentially combine the
different types of electrophoresis and run two-dimensional (2-D) gels. A common form of 2-D
gel electrophoresis is to first separate proteins by IEF in 'tube' gels. These gels are then equili-
brated in SDS gel electrophoresis sample buffer and subjected to SDS-PAGE in a 'slab' gel. (It
is necessary to carry out the IEF first since SDS interfere with the isoelectric focusing.) The 2-
D separation results in higher resolution since proteins are being separated according to two
distinct properties (i.e., charge and size).

Practical Considerations
low ionic strength
protein precipitation
heating
gradient breakdown
98


PROTEIN DETECTION FOLLOWING ELECTROPHORESIS

Total protein stains. Coomassie-blue staining is a
popular method for the detection of proteins following
electrophoresis. The procedure is to 'fix' the proteins into
the gel, usually in methanol and acetic acid, and then to
incubate the gel in a solution containing the dye (see
Appendix 2). Excess dye is removed by destaining in the
acetic acid and methanol solution. The method is cheap and easy to carry out, but of limited
sensitivity.

Silver staining is generally 10-100X more sensitive that Coomassie-blue staining. Three
basic types of silver staining methods have been described. All three methods are based upon
binding silver ions to proteins and reducing the Ag
2+
to metallic silver. The silver interacts with
primary amines or sulfhydryls. In the diamine method, silver diamine complexes are stabilized
with ammonium ions and the free amine groups of proteins. The silver bound to protein is then
reduced to metallic silver with formaldehyde. In the non-diamine method, AgNO
3
reacts with
protein under acidic conditions, followed by reduction to metallic silver with formaldehyde
under alkaline conditions. Photodevelopment relies on light
(photons) to reduce silver ions to metallic silver. Some silver
stain protocols result in certain types of proteins being stained
particular colors, thus aiding in the identification of proteins.

Fluorescence. Fluorescent dyes that bind non-covalently to proteins are also used for
protein detection following electrophoresis. Following incubation with the dye and destaining
the gels are examined under ultraviolet illumination. Polyacrylamide exhibits some background
fluorescence. Proteins can also be convalently labeled with fluorescent probes prior to
Coomassie Blue Stain
Silver Stain
Fluorescence
Autoradiography/Fluorograph
y
Enzymatic
diamine (or ammonical)
non-diamine
photodevelopment
99
electrophoresis. However, attaching fluorescent groups to proteins will change their charges.

Autoradiography and Fluorography. Radioactive proteins can be detected by exposing
X-ray film to the gel. Radioactivity emitted from the proteins will activate silver grains in the
film emulsion which will be converted into metallic silver during development. The reduction
to metallic Ag will result in a dark spot or 'band' appearing on the film in positions which
correspond to proteins. Fluorography refers to the exposure of film to secondary light emitted
by either intensifying screens or fluors upon their exposure to radiation.

Proteins can be made radioactive by incubating cells in the presence of radioactive
amino acids or other compounds that are naturally incorporated into proteins (eg., phosphate,
fatty acids, glycosyl groups, etc.). Non-physiological means can also be used to incorporate
radioisotopes into proteins. Iodination involves the reaction of radioactive iodine (
125
I or
131
I)
with proteins by one of several methods. The iodine in incorporated primarily into tyrosine
residues and to a lesser extent into cysteine and histidine residues. Several reagents are available
that will react with either free sulfhydryls (eg., cysteine) or primary amines (eg., lysine). Such
alkylating agents can be used for the incorporation of either
3
H or
14
C. In most applications the
proteins are radiolabeled before electrophoresis. It should be noted though that alkylating
primary amines will affect the charge of the protein.

Gels are dried before carrying out autoradio-
graphy or fluorography. A gel dryer removes all of
the water and solvent from the acrylamide matrix by
heating the gel while under vacuum. This results in
the matrix collapsing into a thin layer. The dried gel
is placed next to a sheet of X-ray film in a light tight
cassette and exposed for an appropriate amount of
time depending on the isotope and the amount of radioactivity. The exposed film is developed.

Some emissions of high energy isotopes (eg.,
32
P and
125
I) will pass through film
without exposing it. An intensifying screen is placed on the opposite side of the X-ray film
from the dried gel. Emissions passing through the film will then interact with the intensifying
screen which will fluoresce. The light from this fluorescence will then expose the X-ray film.
Using an intensifying screen will usually decrease the exposure time by a factor of 7-10X. The
efficiency of the intensifying screen is highest at -70
o
.

Emissions from low energy isotopes (eg.,
3
H) and some medium energy emissions (eg.,
14
C and
35
S) are not energetic enough to escape the dried gel matrix. In this case a scintillation
fluor is impregnated into the gel before drying. The radioactive emissions then interact with the
fluors resulting in fluorescence. The fluorescence will then expose the X-ray film. Inclusion of
fluors in the gel will not enhance the detection of high energy isotopes, and likewise,
intensifying screens will not enhance the detection of low and medium level isotopes.

Phosphor-imager machines use screens containing storage-phosphors as a replacement
for the use of film. These phosphor-imager screens trap the energy of radioactive emissions and
are sensitive to both beta particles and gamma rays, but insensitive to visible light. The storage-
Autoradiography and Fluorography
Isotope
Energy
Detection
Method
Sensitivity
(dpm/mm
2
)
Screens 2-5 High
(
32
P,
125
I)
Direct 0.5
Fluor 15-25 Medium (
14
C,
35
S)
Direct 2
Low (
3
H) Fluor only 10-20
100
phosphor captures energy with an efficiency around 100% for any particle that strikes the
screen. Subsequent scanning the screen with a laser beam releases the stored energy as blue
light. This blue light is then converted into an image file for display and quantitation by the
phosophor-imager's computer. Such high efficiency makes it possible to detect low levels of
radioactivity after short exposure times. The phosphor screens can be used repeatedly by
'erasing' the screen following scanning.

Enzyme. It is possible to detect enzyme activity following gel electrophoresis of some
proteins. Generally it is necessary to carry out non-denaturing gel electrophoresis or IEF since
proteins will need to be in a native configuration for the expression of activity. However, some
enzyme activities can still be detected following SDS-PAGE. This will require that activity is
located in a single polypeptide chain (i.e., no multi-subunit enzymes) and that the protein is able
to refold into an active configuration following electrophoresis. To promote this refolding the
sample is initially solubilized in a lower concentration (eg., 0.1%) of SDS than normal and is
not heated. Then following electrophoresis the SDS is removed from the gels by washing the
gels in an appropriate buffer. The removal of the SDS and the refolding of the proteins can be
facilitated in some situations by including a non-ionic detergent (eg., Triton X-100) in the wash
buffer. This will lead to a replacement of the SDS by a more approiate detergent.

A common method to detect enzyme activity following
electrophoresis is to use substrates which form insoluble
products. These insoluble products will form a precipitate in the
gel at the position of the enzyme activity. For example,
tetrazolium salts are frequently used as color indicators for the
detection of enzyme systems in which reduction equivalents are
formed. The procedure is to incubate the gel with the
appropriate substrates and the indicator molecules. For exam-
ple, lactate dehydrogenase (see figure) can be detected by
incubating the gel with lactate, NAD
+
, phenazine methosufate
(PMS) and nitro blue tetrazolium (NBT). The NADH formed
by the dehydrogenease is not very efficient at transferring
hydrogens and electrons to NBT and therefore an intermediate
electron acceptor is needed (i.e., PMS). NBT is a soluble
yellow substance which when reduced to a formazan results in
the formation of an insoluble blue compound. Other dehydro-
genase activities or other redox reactions are detected by using the appropriate primary substrate
and changing the intermediate electron acceptor or tetrazolium salt as necessary.

It is also possible to detect enzymes which are generally not considered to be involved
in redox reactions using a substrate that can be coupled to the reduction of tetrazolium salts to
formazan dyes. For example, phosphatase activity can be detected by incubating with 5-bromo-
4-chloro-3-indolyl phosphate (BCIP) and NBT. After dephosphorylation of BCIP by the
phosphatase, the resulting indoxyl reduces the NBT to the foramazan dye. This subtrate pair
is widely used for the detection of enzyme-linked antibodies in immunoblotting procedures
(see chapter on immunoassays).

101
Proteases can also be detected after electrophoresis by co-polymerizing the gels with
gelatin or other proteins. Following electrophoresis and removal of the SDS, the gel is incu-
bated under conditions which promote proteolysis and then stained with Coomassie blue. Clear
regions will denote the positions of proteases.

Quantification. The results obtained from gel electrophoresis tend to be qualitative and
difficult to precisely quantify. One possible way to quantify the amount of a specific protein is
to electrophorese known amounts of a protein and to subjectively compare the staining
intensities with that of the unknown protein. The stained protein bands could also by excised,
the dye extracted with a solvent and the amount dye determined with a spectrophotometer. The
absorbance values could then be used to generate a standard curve. However, individual
proteins bind different amounts of Coomassie blue and there will be gel-to-gel variations.
Similarly radioactive proteins can be excised from the gel, the gel slice solubilized and the
amount of radioactivity determined by liquid scintillation counting. This method can be rather
laborious if all the proteins within a sample are being evaluated.

Scanning densitometry is usually a convenient method to quantify the amount of a
particular protein. Gels, autoradiographs, or even photographs are scanned and the peak height
and areas of bands are determined. The values are then used to calculate the amount of protein
based on standard curves. The uniformity of the shape and width of the bands also needs to be
considered.

PREPARATIVE ELECTROPHORESIS

Gel electrophoresis is a high resolution technique and can potentially be used for protein
purification. One major problem, however, is the limited amount of protein that can loaded onto
a gel. Another major problem is the recovery of proteins from gels, especially if the gels have
been fixed and stained. One possible method is by simple diffusion. The excised gel slice is
resuspended in a buffer and the protein allowed to diffuse out. This process is slow and the
protein is excessively diluted. Alternatively, the protein can be recovered from the gel slice by
electroelution using commercially available apparatuses. Both diffusion and electroelution are
characterized by low yields. An alternative strategy is to transfer the protein to a membrane
support and analyze the protein bound to the membrane as is done in immunoblotting or micro-
sequencing.

If the goal is to raise antibodies against the purified protein then it may not be necessary
to remove the protein from the gel. The excise gel slice containing the protein of interest is
frozen in liquid nitrogen and ground into a fine powder with a mortar and pestle. The ground
gel is then resuspended in saline and used to directly immunize animals. The protein is slowly
released from gel slices and actually enhances immunization through the 'depot' effect.

Commercial apparatuses for preparative electrophoresis are available. Most treat the gel
as a column and continue electrophoresis until all of the proteins migrate off from the bottom of
the gel. At the bottom of the gel is a chamber sealed off with a dialysis membrane to prevent the
proteins from migrating into the tank buffer. The protein coming off from the bottom of the gel
are continuously eluted and pumped to monitors and fraction collectors.
102



A commercially available preparative IEF apparatus, the Rotofor

(Bio-Rad), avoids the

problems with gels by conducting the focusing in free solution. The apparatus consists of
column (i.e., electrophoresis chamber) which is held in an horizontal position. The electrode
solutions are separated from the electrophoresis chamber by ion exchange membranes and
gaskets. A ceramic cooling core runs down the center of the electrophoresis chamber and the
chamber is rotated around the horizontal axis during electrophoresis to dissipate heat, minimize
convection, and maintain efficient electrical contact. The chamber is filled with the protein
solution containing ampholytes and usually sucrose to the increase density. The mixture is
subjected to electrofocusing by applying the appropriate voltage. The electrophoresis chamber
is partitioned into 20 chamber each divided by a porous polyester screen with a pore size of
approximately 6 m. These screens offer some resistance to fluid convection by do not impede
the flow or current or the movement of proteins. After the power is turned off the 20 fractions
are collected with a special device that rapidly harvests the 20 fractions before substantial
mixing occurs between them.

103

104
ELECTROPHORESIS APPENDIX 1. SIZE CALCULATION.

Caluculation of molecular mass from SDS gels.

Molecular masses of proteins can be estimated by comparing the migration of proteins of
interest to standards of known size. The relative mobilities of the standards are plotted
against the log of their molecular masses. The sizes of unknown proteins are then
extrapolated from the standard curve.

The relative mobility (or R
f
) is calculated by dividing the distance
the protein migrate by the total length of the gel or the dye front as
illustrated by the figures and table.

mass
(daltons)
log
(mass)
distance
migraged
(mm)
Rf ( 101
mm)
205,000 5.31 14.5 0.14
116,000 5.06 27 0.27
97,000 4.99 33 0.33
66,000 4.82 45 0.45
45,000 4.65 67 0.66
29,000 4.46 92 0.91




4.40
4.60
4.80
5.00
5.20
5.40
0.00 0.20 0.40 0.60 0.80 1.00
Mobility (R
f
)
L
o
g

(
m
a
s
s
)

105
ELECTROPHORESIS APPENDIX 2. GEL STAINING


Coomassie Blue Staining Following Electrophoresis

G-250 Method (quick)

1. Fix the proteins by incubating the gel in the destain solution (10% methanol and 7%
acetic acid) for at least 15 minutes. Gels can be left in fixative indefinitely.

2. Briefly rinse the gels with water.

3. Stain the gels for 20 minutes with 0.1% Coomassie-blue G-250 in 3.5% perchloric acid.
Gels thicker than 0.75 mm may need to be stained longer.

4. Destain the gel with several changes of the destain solution (10% methanol and 7%
acetic acid) over the next 30-60 minutes. (Works best if the first 1-2 washes carried out
in the first 5-10 minutes.)

R-250 Method (slow)

1. Fix the proteins by incubating the gel in the destain solution (10% methanol and 7%
acetic acid) for at least 15 minutes. Gels can be left in fixative indefinitely.

2. Stain the gels for 3 hours with 0.1% Coomassie-blue R-250 in 50% methanol + 10%
acetic acid.

3. Destain the gel with several changes of the destain solution (10% methanol and 7%
acetic acid) over the next 6-24 hours.

Notes

For optimal results, the G-250 should be added to the perchloric acid. Water should then
be slowly added while stirring. Continue stirring for at least one hour and then filter the
solution over filter paper.

The R-250 solution should also be filtered after its preparation.

Other concentrations of methanol and acetic acid also work in the fixative and/or
destaining solutions

107

CHAPTER 12--PROTEIN PURIFICATION OVERVIEW

There are various methods that separate proteins
according to distinct chemical and physical properties.
Proteins can be purified by sequentially applying these
techniques. Before purifying a protein one needs to
consider how much protein is needed, how pure it needs
to be, whether the purified protein needs to be in its
native configuration, or express an activity. These
considerations will be determined by the ultimate use for
the purified protein (Box). If the purified protein is going
to be used in functional and structural studies, isolating the protein in its native and active
configuration will be important. However, if the goal is to determine a partial amino acid
sequence then concerns about denaturation are minimal.

The optimal method to purify a protein is empirically determined by carrying out small-
scale pilot studies. After the optimal conditions for a particular purification step is worked out,
then the procedures can be scaled up. The order in which the steps are carried out will also
affect the final outcome. In general, though, it is best to start with high capacity methods that are
easily and quickly carried out and then proceed to high resolution/low capacity methods.

Capacity refers to how much total protein can be readily accommodated and resolution
refers to the ability to separate the protein of interest from contaminating proteins. Generally, as
the resolution of the technique increases, the time required to carry out the technique increases
and the protein capacity decreases (Box). One exception is that gel filtration is not usually a
high resolution technique, as well as being low capacity and relatively difficult to carry out. The
capacity and resolution of affinity chromatography depends on the ligand. Some ligands afford
both high capacity and high resolution. The resolution of any of the chromatographic technique
can be substantially enhanced through the use of HPLC. However, the capacity of HPLC
columns is much lower so that it may not be useful until later in the purification protocol.

In general:
Method
differential
solubility
ion exchange
hydrophobic
electrophoresis
capacity
decreases
resolution
increases
time & effort
increase
deviations
gel filtration (low capacity, low resolution)
affinity (depends on ligand)
HPLC/FPLC ( resolution, capacity)

Why purify proteins?
detailed studies on function
determination of structure
industrial/pharmaceutical
applications
generate antibodies
amino acid sequence
determination
108
The time and number of manipulations should be minimized in designing purification
protocols. For example, the order of the separation techniques can be chosen as to minimize the
number of buffer changes when other factors are equal. Another way to expedite purification is
to exploit a unique feature of the protein. In this regard, affinity chromatography exploits the
specificity of ligand binding to produce high resolution separations in a single step.

An example of exploiting a unique feature is the purification of calmodulin on hydro-
phobic columns in the presence of Ca
2+
. Binding of Ca
2+
causes a conformation change in
calmodulin so that a hydrophobic domain is exposed on the protein's surface. The function of
this hydrophobic domain is to bind to other proteins that are regulated by calmodulin. In the
absence of Ca
2+
, calmodulin is hydrophillic and does not bind to other proteins. Hydrophobic
chromatography in the presence of Ca
2+
results in absorption of calmodulin to the column.
Specific elution of calmodulin is accomplished by the addition of EGTA which chelates the
Ca
2+
resulting in a conformation change in calmodulin and the loss of its surface hydro-
phobicity. The surface hydrophobicities of the other proteins are not affected by the presence or
absence of Ca
2+
and therefore do not elute with EGTA. the activity of the bound protein. binds
to hydrophobic columns, but does not bind in the absence of Ca
2+
. This unique feature permits a
rapid and nearly complete purification of calmodulin in a single step.

Another example of exploiting unique
features is the use of gel filtration in the purifi-
cation of proteins which can be reversibly
dissociated subunits. For example, gel filtration
can be carried out in the presence of low salt
(typically 0.15 M NaCl) and the protein will
elude from the column corresponding to the
molecular weight of the complex. If the complex
can be reversibly dissociated by high salt (eg., 1
M NaCl) and re-subjected to gel filtration chromatography, it will elude from the column
corresponding to its subunit molecular weight. The intact protein can then be reformed by
removing the salt. These sequential gel filtration columns in low and high salt will result in
substantial purification with a normally low resolution technique.

Concentration. Many chromatographic proce-
dures result in a dilution of the protein sample. To
avoid the possible negative effects of dilute protein
solutions, it is often necessary to re-concentrate the
sample between chromatographic steps or after the
purification is completed. Precipitation is a convenient method in cases where (NH
4
)
2
SO
4
does
not adversely affect the protein or next chromatographic step. Water can be completely
removed by lyophilization, or freeze-drying. However, many proteins are not stable following
lyophilization and salt concentrations will be dramatically increased. Water can also be
removed by placing the protein solution in a dialysis bag and surrounding it with solid high
molecular weight polyethylene glycol. Water and buffer components will diffuse out of the
dialysis and the ionic strength remains approximately constant. Similarly, dialysis against a
buffer solution containing 50% glycerol will concentrate the solution 4-5 fold as the water
lyophilization (freeze drying)
precipitation (eg., (NH
4
)
2
SO
4
)
dialysis against solid PEG
dialysis against 50% glycerol
ultrafiltration

109
diffuses out of the dialysis bag. This may be especially useful as a terminal purification step in
that glycerol is often added to purified proteins to increase the stability.

Ultrafiltration involves forcing a pro-
tein solution through a membrane with a
defined molecular weight cutoff. The protein
does not pass through the membrane, whereas
the other buffer components do resulting in
no change in the ionic strength. This method
is usually relatively rapid and does not
adversely affect the protein sample. Ultrafiltration is either carried out in a special apparatus or
by centrifugation using a disposable filter cartridge that fits into the centrifuge tube. The protein
sample is placed into the cartridge and centrifuged until the protein solution, which remains
above the filter, is at the desired volume. The filtrate can also be recovered and evaluated. A
wide range of sizes are available for both the ultrafiltration apparatus and the centrifuge
cartridges.

Evaluation. It is important to monitor and evaluate the protein of interest during the
purification procedure. Two important criteria to monitor are fold-purification and percent
yield (see Appendix). The percent yield refers to the amount of the protein recovered following
a purification step and is determined by dividing the amount of total activity recovered by the
total activity in the starting material. The total activity of the starting material is determined by
multiply the activity/ml by the total volume of the sample applied to the column or other
separation method. The fractions containing the protein of interest are then pooled and the
activity/ml is determined and multiplied by the total volume of the pooled fractions. Generally
steps that result in low yields should be avoided unless the protein is available in large amounts.

The fold-purification is determined by dividing the specific activity following the
purification procedure by the specific activity before the procedure. The specific activity is the
activity per mg protein. The specific activity is determined by dividing the activity/ml by the
protein concentretration (mg/ml) in both the starting material and the pool fractions following
the separation technique. After each step the specific activity should go up since unwanted
proteins are being removed. This increase in specific activity provides a measure of the degree
of purification obtained by that method(s). Another way to view fold-purification is the removal
of unwanted proteins. For example, a 4-fold purification means that 75% of the contaminating
proteins were removed by that step. Both the percent yield and fold-purification can be
determined after each step or with the original starting material to evaluate the entire
purification process.

In addition, the qualitative aspects of the purification should be analyzed by gel electro-
phoresis. A protein purified to homogeneity should exhibit a single 'band' when analyzed by
non-denaturing electrophoresis. The number of polypeptides (i.e., bands) observed after SDS-
gel electrophoresis will depend upon the subunit composition of that protein.


110
MICROSEQUENCING AND PEPTIDE MAPPING

A partial amino acid sequence can provide information about a proteins identity or be
used in the design of recombinant DNA probes. Through chemistry know as the Edman degra-
dation cycle the N-terminal amino acids can be sequentially removed and then determined.
Generally instruments which automate this sequential removal of amino acids and their analysis
are utilized. Generally proteins or peptides to be sequence are submitted to a service provider
and sequenced for a fee depending on the number of residues sequenced.

Edman degradation chemistry. Protein samples are usually absorbed or coupled to a
membrane or glass fiber filter support which is placed in the reaction chamber of the instrument.
Each cycle of the Edman degradation cycle consists of three steps: coupling, cleavage and
conversion. The chemicals necessary for each of the steps are sequentially pumped into the
reaction chamber. In the coupling step, the unmodified N-terminus of the peptide is reacted with
phenyl isothiocyanate to form a phenylthiocarbamoyl-peptide. This is followed by addition of
either liquid or gaseous trifluoroacetic acid which will cleave PTC derivatized N-terminal
amino acid from the peptide resulting in the formation of an anilinothiazoline-amino acid
derivitive. This unstable ATZ-amino acid is pumped from the reaction chamber into a
conversion chamber where it is converted to a more stable phenylthiohydantoin derivative. The
immobilized peptide remains in the reaction chamber and the N-terminus of the next residue is
available for another round of Edman degradation. The PTH-amino acid is transferred from the
conversion chamber to an HPLC and the specific amino acid is determined from its elution
time. Depending on the quantity and quality of the sample up to 30 cycles can be carried.
Generally about one pmole of amino acid can be detected although more sensitive instruments
are available.

Microsequencing following gel electrophoresis. The N-termini of proteins can be
sequenced following separation by gel electrophoresis. This obviates the need to purify the
protein to near homogeneity by more laborious conventional methods and allows one to take
advantage of the high resolution of SDS gel electrophoresis. The protein of interest should not
significantly co-migrate with contaminating proteins and be at least 80% pure. Generally 10-
100 pmoles of the protein, which corresponds to 0.5-5 g of a 50 kDa protein, will be needed to
make carry out the analysis. If no sequence data is obtained from 100 pmoles of protein then it
is generally assumed that the N-terminus is blocked (see below). Whereas if less than 10 pmoles
of sample are used it may be difficult to determine if the lack of sequence data are the result of a
blocked N-terminus or due to insufficient material.

Following gel electrophoresis (either 1- or 2-dimensional) the proteins are electrophore-
tically transferred from the gels to a membrane support. The membrane of choice is a modified
PVDF membrane which has good protein retention and is able to withstand the solvents, acids
and bases used in the Edman degradation chemistry. The protein of interest is detected by a
brief staining with Coomassie blue (or other dyes) and the band excised. Components of the
electrophoresis and transfer buffers, such as salts, Tris, glycine, SDS and acrylamide, interfere
with the sequencing reactions and must be completely removed by extensive washing in water.
The membrane piece containing the protein to be sequence is submitted to the sequencing
service.
111

Sequencing internal peptides. Edman chemistry is dependent on a free amino group at
the N-terminus. Apporximately half of eukaryotic cellular protein are modified (eg., acylation)
at the N-terminus and do not have a free amine in this position (i.e., 'blocked' N-terminus). In
addition, sometimes during the purification and preparation of proteins for N-terminal
sequencing the proteins can become blocked. There are some methods (chemically and
enzymatically) to remove the N-terminal blocking group of a protein or peptide. These include
deformylation, deacetylation and removal of the pyroglutamate group. The choice of a
particular method depends on the protein or peptide and the nature of the blocking group.
However, in most cases because of the uncertain nature of the blocking group an internal
peptide sequence is determined.

Proteins can be treated with chemicals or
site specific proteases to generate peptide frag-
ments. For example, CNBr cleaves proteins on the
C-terminal side of methionine residues. Many
protease recognized specific amino acid residues
and reproducibly cleave proteins at specific sites.
The resulting polypeptides are then separated
(typically on HPLC reverse phase columns). The
peaks corresponding to peptides are collected and absorbed onto glass fiber filters as a solid
support. Reverse phase chromatography is especially advantageous in this application since it is
carried out in volatile solvents. The glass fiber filters containing the peptides are submitted to
the service provider and the N-termini sequenced.



Protocols and discussions about preparing proteins for microsequencing can be found at
http://www.biol.berkeley.edu/crl/microchemical/protein_sequencing.html.

Peptide mapping. It is also possible to use site specific proteases to compare proteins
and determine if they are related. The general procedure is to digest a protein with a protease
and characterize the pattern of peptides formed. Each protein will exhibit a unique pattern of
peptides. Such procedures are known as 'protein fingerprinting' or 'peptide mapping'. The
peptides can be analyzed by HPLC, thin-layer chromatography (TLC), or gel electrophoresis. In
two-dimensional peptide mapping the peptides are separated by TLC followed by high voltage
electrophoresis followed by separation by high-voltage electrophoresis in a second dimension.

Another form of peptide mapping involves excising a protein band from the gel and
Site Specific Cleavage of Proteins
chemical CNBr Met-X
skatole Trp-X
trypsin Arg-X
Lys-X
proteolytic
V8 Glu-X
Asp-X
112
loading it on a second gel of higher acrylamide concentration. A protease is added to the well
and the protein of interest is partially digested during the re-electrophoresis. Proteins will
exhibit a distinct pattern of smaller polypeptides. This method, often referred to as the
Cleveland method, is rapid and easy to carry out and allows proteins to be directly compared in
neighboring lanes. The pattern of polypeptides can be used to assess whether two proteins are
similar or different.

Protein modification. Sometimes it is necessary to chemically modify proteins. and in
particular to radiolabel proteins. Several methods are available to directly couple
125
I to proteins.
Tyrosine residues and to a lesser extent histidine and cysteine, are labeled. Free amines, as
found on lysine and arginine residues are quite reactive and numerous aldehydes, esters and
imidoesters are available for modifying proteins. The free sulfhydryl group of cysteine is also
reactive and easy to alkylate. Alkylation prevents the formation of disulfide bonds. Bifunctional
reagents can also be used to cross-link proteins.


APPENDIX. FOLD PURIFICATION AND RECOVERY

Example of a form to record data on the purification of an enzyme.

Protein Activity

Purification Procedure


Vol. l g mg/ml total units/ml total sp. act.


%yield

fold-
purify









The effectiveness of a purification procedure is evaluated by monitoring the recovery of the protein of interest and its relative
purification after each step. The volume, protein concentration and amount of activity should be determined for the crude homogenate
and all subsequent manipulations. The % yield for any single procedure (i.e., recovery) is determined by dividing the total activity
recovered in the pooled fractions by the total activity in the preceding step. The overall % yield is determined by dividing the total
activity by the total activity in the crude homogenate (i.e., the first step). The total activity is determined by multiplying the total
volume by the units/ml. The specific activity (units/mg protein) represents the proportional amount of the protein of interest and should
go up with each purification step since the total protein concentration is decreasing. The fold-purification is determined by dividing the
specific activity after a particular step by dividing the specific activity by the specific activity of the preceding step or the specific
activity of the crude homogenate. The maximum fold-purification that can be obtained is determined by the percentage of the protein
of interest in the crude homogenate. For example, a protein that represents 0.1% of the total cellular protein should be homogeneous
after a 1000-fold purification.
114




PART III



Immunological Methods



Topics covered:

Antibody Production
Monoclonal Antibodies
Immunoassays
115
CHAPTER 13--IMMUNIZATION

Antibodies are proteins that specifically recognize and bind to other molecules with a
high affinity and specificity. This high degree of specificity makes antibodies powerful reagents
for the characterization of proteins and other macromolecules. Antibodies are generated against
specific proteins or other macromolecules by immunizing animals and collecting the sera. The
immune sera contains a mixture of antibodies that recognize various epitopes on the antigen.
The immune system can also respond to foreign substances via a cellular immune response.

IMMUNOGENICITY

Immunogenicity (i.e., the level of the immune response) depends on the chemical nature
of the immunogen and the ability of the animal to respond to the immunogen. An immunogen is
a substance that induces a specific immune response, whereas an antigen is defined as a
substance that reacts with the products of a specific immune response. An antigen can be an
immunogen if it is correctly presented to the immune system. In addition, there is a genetic
component to immunogenicity in that some substances are immunogenic in one species but not
in another. Similarly, some substances are immunogenic in one individual but not in others (i.e.
responders and non-responders). Immunogenicity can be manipulated, though, by either
modifying the immunogen or stimulating the animal.

An immunogen generally needs to be a foreign sub-
stance since the immune system normally discriminates
between self and non-self. In addition, immunogens need to
have significant chemical complexity and be of a large size
(eg., > 10 kDa). Proteins and polysaccharides tend to make
good immunogens, whereas nucleic acids and lipids are
usually poorly immunogenic. Lipids can function as
haptens, though (see below).

Optimal immunogens should also have B-cell epitopes, T-cell epitopes and class II
binding sites (Box). Epitopes are the structural determinants recognized by either antibodies or
T-cell receptors. B-cell epitopes bind to membrane bound antibodies on the surface of B-cells
and stimulate the production and secretion of antibodies. This can occur independent of T-cells,
but the antibody response (both in amount and affinity) tends to be lower. Polysaccharides are
generally T-independent immunogens in that they are characterized by the repetition of the
same antigenic determinant (i.e., the saccaride subunits) and are somewhat resistance to
degradation. B-cell epitopes in proteins tend to be surface exposed hydrophilic residues and
usually consist of 4-8 amino acid residues

T-dependent immunogens are characterized by a few copies of a complex mixture of
epitopes as exhibited by proteins. T-dependent antigens are also processed into small peptides
and expressed on the surface of antigen-presenting cells via the MHC class II proteins. For
example, macrophages will take up foreign organisms or particulate matter and degrade these
substances with hydrolytic enzymes in the lysosomes. Immunogens in particulate form are
better since they are more efficiently phagocytosed by macrophages. Proteins will be
foreigness
molecular complexity
molecular size
B-cell epitope
T-cell epitope
class II binding site
degradation/presentation
particulate/phagocytosis
116
proteolytically processed into peptides, some of which will bind to the class II MHC molecules
and transported to the cell surface. Thus a good immungen needs to contain peptide sequences
which bind to the MHC protein. This also explains genetic restrictions in the immune response
since individuals have different repertoires of MHC proteins and will react differently to
immunogens.

Potential Problems Associated with
Immunogens and Possible Remedies
Property Defect Result Remedy
Lacks B-cell
epitope
No B-cell
recognition
No response None
Lacks Class II
site
No presentation No response or only
primary response
Conjugate with class II
site or switch animals
Lacks T-cell
epitope
No T-helper
involvement
No response or only
primary response
Conjugate with T-epitope
or switch animals
Non-degradable No presentation No response or only
primary response
None
Small size No T-helper
involvement
No response Conjugate with carrier
(hapten)
Non-particulate Poor
phagocytosis
Weak response Self-polymerize or
couple to beads

A macrophage with MHC associated
peptides on its surface is an antigen presenting
cell. Antigen presentation is not restricted to
macrophages and other types of cells can also
function as antigen-presenting cells. In the
MCH-bound state the peptides (i.e., T-cell
epitope) are recognized by a T-cell receptor
found on the surface of T-lymphocytes. T-cell
receptors exhibit a high degree of specificity in
regards to the peptide sequence recognized and
only those T-cells expressing the correct recep-
tor will be activated. The exact immune re-
sponse will depend in part on the nature of the
activated T-cells. For example, type 2 T-helper
cells (T
h
2) will secrete cytokines that stimulate
B-cells in the presence of antigen to replicate
and then differentiate into the antibody-producing plasma cells. Some of the activated B-cells
will differentiate into memory cells. Upon a second encounter with the immunogen memory
cells more rapidly differentiate into plasma cells resulting in a faster and more intense antibody
response. Thus, T-dependent immunogens are also characterized by 'immunologic memory' and
a strong secondary response (i.e., boosting). Immunogens lacking class II sites or T-cell
epitopes can be conjugated to good MHC-binding sites or T-cell epitopes.

IMMUNOGEN PREPARATION

117
The preparation of the immuogen and the immunization protocol can affect the nature of
the antibody produced. How the antibody will be used in subsequent assays and the goals of the
experiments need to be considered before purifying the antigen and immunizing animals.
Typically a purified protein is used as an immunogen. A major decision concerns how pure the
antigen needs to be before starting the immunization schedule. Ideally an immunogen should be
purified to homogeneity for the production of monospecific antisera. Partially purified or crude
immunogens can also cause problems with suppression and 'antigenic competition' in that some
proteins are more immunogenic than others. This can result in a strong immune response
against a minor contaminant in the preparation.

Synthetic peptides and recombinant proteins can be used in situations where it is not
possible to purify sufficient protein of the desired purity. In addition, the use of recombinant
proteins or synthetic peptides will permit the production of antibodies against particular
epitopes or domains of the protein. Synthetic peptides or other small chemicals, often called
haptens, need to be coupled to a carrier, or larger protein, before immunization. Conjugation to
a carrier ensures an adequate immunogen size, increased haptenic epitope density, and source of
heterologous T-cell epitopes. The haptens or peptides are chemically cross-linked to the carrier.
The recombinant DNA approach (i.e., genetic engineering) is also amenable to the inclusion of
good T-cell epitopes and class II MHC binding sites.

The structure of the immunogen is
another factor to be considered since the
nature of the epitope recognized by an anti-
body will affect its performance in the various immunoassays. Immunization with non-
denatured (i.e., native) proteins tends to produce antibodies against conformation epitopes,
whereas immunization with denatured proteins tends to produce antibodies against sequential or
linear determinants. Native proteins tend to be more immunogenic than denatured proteins. But
it is generally easier to obtain a higher degree of purity under denaturing conditions.

In cases where antibodies that recognize denatured protein are satisfactory, SDS gel
electrophoresis can be used as the final step in the purification protocol. The protein band of
interest is excised from the gel and minced into small pieces (~1 mm
3
). The gel pieces are
placed into a mortar, frozen with liquid nitrogen, and ground into a fine powder with the pestle.
A slurry is formed by resuspending the gel in saline and used for immunization. Protein will be
slowly released from the gel and result in a depot effect and therefore adjuvants are generally
not needed.

IMMUNIZATION

Choice of Animal. A wide range of vertebrate species can be used for the production of
antibodies (Table), but the two most often used animals are mice and rabbits. Rabbits react to
most immunogens and relatively large quantities of antisera are obtained. The disadvantages of
rabbits are that larger quantities of antigen are needed and rabbits often exhibit a high
background response to heterologous antigens. In particular, high levels of antibodies to E. coli
antigens will complicate the analysis of antibodies against recombinant proteins.

Native vs Denatured
conformation epitopes sequential epitopes
more immunogenic more highly purified
118
Animal Advantages Disadvantages
Rabbits large amounts of sera
strong antibody response
easy to maintain
require more immunogen
high background
Mice/Rats inbred strains
monoclonal antibodies possible
easy to maintain
small amounts of sera
Guinea pigs generally high titer antibodies
easy to maintain
difficult to bleed
small amount of sera
Goats/Sheep very large amounts of sera difficult to maintain
Chickens good for some highly conserved
mammalian antigens
difficult to maintain

Mice are easy to handle and several different genetic strains are available. The different
genetic strains provide a more uniform response to immunogens and also some control of the
immune response since the different strains may respond differently to the same antigen. In
addition, mice are used models for many infectious diseases. The major disadvantage of mice is
that a small amount of antisera is obtained.

Adult animals should be used for immunizations (i.e., rabbits 6-10 lb. and mice > 7
weeks). If possible, 2-3 rabbits and 5-10 mice should be immunized with the same immunogen
to control for animal-to-animal variation.

Main Classes of Adjuvants
Adjuvant Comment
aluminum
salts
The immunogen is absorbed to a gel-like precipitate of aluminium hydroxide, aluminium
phosphate, or alum. Action is due to efficient uptake by macrophages and slow release.
water-in-oil
emulsion
Aqueous microdroplets containing the immunogen and stabilized by a surfactant are in a
continuous oil phase such as mineral oil or squalene. Immune modulators (eg.,
mycobacteria, MDP) are often incorporated into the emulsion. The emulsion provides a
good short term depot effect and is especially useful for soluble immunogens.
oil-in-water
emulsion
The immunogen is associated with the surface of microdroplets of oil (eg., sqalane) in a
continuous aqueous phase and stabilized by surfactants (eg., Tween 80). Highly suited
for lipophilic or amphipahtic immunogens since it is important for the immunogen to be
incorporated into the oil phase.
saponin Saponins form complexes with proteins and lipids resulting in an open cage-like matrix
(i.e., micelles) which are sometimes referred to as immunostimulating complexes
(ISCOM).
liposomes Immunogens can be incorporated into or absorbed to liposomes (single or multilayered
bilayer vesicles) composed of phospholipids and cholesterol. Immune modulators such
as LPS and MDP can also be incorporated.

Adjuvants. Adjuvants are substances that enhance the overall immune response. Most
adjuvants contain two components. One component will form a deposit that prevents the rapid
breakdown of the antigen and promotes a slow release of the antigen, called the 'depot effect'.
There are several different classes of adjuvants with distinct mechanisms by which the depot
effect is evoked (Table).

119
The second component of many adjuvants non-specifically stimulates the immune
system (mitogenic or polyclonal activation). For example, killed bacteria, especially mycobac-
teria, or bacteria components are included in some adjuvants. Common bacterial components
included in adjuvants are lipopolysaccharide (LPS) or muramyl didpeptide (MDP), a synthetic
analogue of the adjuvant-active component from the cell walls of mycobacteria. These sub-
stances stimulate macrophages and other immune effector cells to release cytokines which then
enhance the antibody response. The immunomodulatory component may also influence the
class of immunoglobin produced as well as the isotype of IgG.

(Review of adjuvants: Cox and Coulter. 1997. Adjuvants a classification and review of their
modes of action. Vaccine 15, 248-256.)

Injection. It is possible to immunize by se-
veral different routes (Table). The route of injection
can also affect the antibody response, both in terms
of quantity of antibody produced and in the qualita-
tive aspects of the antibody. The volume of immuno-
gen and how quickly the immunogen should be
released into the lymphatics or circulation will
influence the choice of injection method. Rabbits are
generally immunized subcutaneously at multiple
sites to stimulate regional lymph nodes. Immunization of mice with large volumes can only be
accomplished intraperitoneally. Intramuscular and intradermal injections are best for slow
release. However, intradermal immunizations are difficult to do correctly. Intraveneous
immunization is only used for boosting (see below) and immunogens containing adjuvants
should not be given intravenously.

The antigen dose will depend on the animal, the purity and the immunogenicity of the
antigen. In general, 0.05-1 mg and 5-50 g of pure antigen are needed for rabbits and mice,
respectively. To obtain the optimal antibody response (i.e., highest titer and affinity) the animals
will need to be reimmunized, or boosted. Boosting promotes class switching (eg., IgM IgG),
higher levels of antibody, and an increase in the affinity of the antibody for the antigen.
Typically, the boosting is carried out at intervals of 2-6 weeks and 3-4 total immunizations are
given.

Initial contact with a new antigen may evoke a primary response. The primary response
is characterized by a initial rise in IgM followed by IgG (Figure). Antibodies following the
primary response are generally transient and will disappear from the circulation. Subsequent
exposure (eg., booster injection) to the same antigen results in a much faster antibody response
as well as significantly higher levels of antibodies which are predominantly IgG. Furthermore,
the antibodies associated with this secondary response persist for a much longer time in the
serum. This additional exposure to the antigen also results in a selection for plasma cells
producing antibodies with a higher affinity for the antigen. This affinity maturation results in
antibodies which react more strongly with the antigen.

Route Comments
subcutaneous sc easy injections
intramuscular im slow release
intradermal id difficult injections,
slow release
intravenous iv only for boosting,
no adjuvants
intraperitoneal ip easy injections,
common in mice
120



COLLECTING AND PROCESSING BLOOD

Serum samples are typically collected 4-10 days after an immunization and tested for
antibody titer. Clotting factors interfere with many assays, and therefore, serum is usually
preferable to plasma (total non-cellular component of blood). Serum is prepared by collecting
the blood without anticoagulants and allowing the blood clot for at least 30-60 min at 37
o
.
Blood clotting can also be performed for 2-4 hours at room temperature or overnight at 4
o
.
Incubation overnight at 4
o
will usually produce more hemolysis. The clot is removed and the
sample centrifuged to remove any remaining blood cells. In some applications it is also
beneficial to inactivate complement by heating the serum at 56
o
for 30 minutes.

ANTIBODY PURIFICATION

Whole sera can be used for most immunological assays and further purification of
antibodies is generally not necessary. Antibodies can be concentrated and partially purified by
precipitation with 40-50% saturated (NH
4
)
2
SO
4
. This will remove a substantial portion of serum
albumin--the major component of serum. Higher levels of purity can be obtained after anion
exchange chromatography on DEAE. The immunoglobulins elute from the column before
serum albumin. Affinity chromatography using protein A or G can also be used. Protein A and
G are bacterial proteins that bind IgG. The various IgG subclasses from different animals
exhibit a range of affinities for these proteins and can be differential eluted from the affinity
columns at different pH values.

121
Specific antibodies can also be purified from polyclonal sera. For example, affinity
purification using the antigen as ligand can be used to isolate the antibodies specific for that
protein. Monospecific antibodies can also be generated by taking advantage of the high
resolution of gel electrophoresis (see Figure below for example). (See section on
Immunoblotting for a detailed discussion of the methods.) It is also possible to preabsorb serum
to remove unwanted antibodies and to reduce background. Examples include E. coli proteins or
the fusion partner in recombinant proteins.




123
CHAPTER 14--MONOCLONAL ANTIBODIES

Serum contains a mixture of antibodies recognizing many different antigens. Further-
more, different epitopes of any particular antigen are also recognized. These non-specific
reactions and cross-reactivity may limit the usefulness of antisera. This heterogeneity is the
result of antibodies being the products of clonal lineages of mature B-cells (see appendix for B-
cell differentiation). Each clonal lineage of a B-cell produces a specific antibody that recognized
a specific epitope. It is not possible to isolate and grow B-cells in culture.

In 1975 Kohler and Milstien described a technique that allowed the growth of clonal
populations of cells secreting antibody of defined specificity, for which they received the 1984
Nobel prize in medicine. The method consists of fusing B-cells with myeloma cells (a type of
B-cell tumor). The fused cell is known as a hybridoma. This hybridoma now has the ability to
grow in culture and it secretes an antibody of defined specificity (phenotype acquired from the
B-cell). The antibody produced by the hybridoma is referred to as a monoclonal antibody
(mAb).

The major advantage of a mAb is its unlimited
supply (Box). The hybridoma cell line has indefinite life
whereas only a limited amount of serum can be obtained
from immunized animals. In addition, the mAb is a
defined reagent that recognizes a single epitope. Indivi-
dual animals may respond differently to the same antigen
resulting in potentially different polyclonal antibodies against the same antigen. Another
advantage is the ability to produce specific antibody with impure antigen. A disadvantage of
mAbs is that only single epitope recognized and the mAb may not have desire specificity or
affinity. Secondly, mAb are time consuming (6-9 months) to produce and requires tissue culture
facilities. Following immunization there are several steps (Box)
involved in the production of monoclonal antibodies.

IMMUNIZATION

Immunized animals are needed for the first step in mAb
production (Box). The same basic factors (i.e., immunogen prepara-
tion, adjuvants, etc.) for the generation of polyclonal sera (see chapter
on Immunization) apply to the production of mAbs. However, rodents, and especially mice, are
almost always the animal of choice. This is primarily due to the availability of mouse myeloma
cell lines, but human and simian hybridomas have been successfully produced by transforming
the lymphocytes with Epstein-Barr virus. If future plans include the production of ascites
tumors then the mouse strain used for the immunization should match the mouse strain from
which the myeloma cells were derived. In some situations it is not necessary to use highly
purified immunogens. For example, it is also possible to use animals immune to an infectious
disease or to immunize with crude antigens. Monoclonal antibodies with the desired
specificities are then selected during the screening stage.

Advantages of mAbs
unlimited supply
defined reagent
Disadvantages of mAbs
single epitope
time consuming
1. Immunization
2. Fusion
3. Selection
4. Screening
5. Cloning
6. Production
124
The mice are immunized, boosted and tested until an antibody response of the desired
specificity is elicited. Generally the monoclonal antibodies obtained will be reflective of the
antibodies found in the polyclonal sera. Therefore it is critical that the mouse exhibits antibodies
of the desired specificity before proceeding with the fusion.

FUSION

Several myeloma cell lines are available. A common
mouse cell line used for the production of mAbs is X63Ag8.653.
The parental cell line was initially derived from a tumor induced
by injecting mineral oil into a Balb/c mouse. The resulting cell
line was subjected to mutagenesis to produce a hypoxanthine-
guanine phosphoribosyl transferase (HGPRT) deficient cell line.
Further cloning yielded X63Ag8.653 which no longer secretes
its own antibody. Therefore hybridomas prepared with this cell line will only secrete antibodies
obtained from the spleen cells.

B-cells are isolated from the immunized mouse. The mouse exhibiting the desired
antibody response is usually boosted 3-4 days before doing the fusion to stimulate the B-cells.
The spleen is most common source of B-cells, but it is also possible to use lymph nodes.
Spleens or lymph nodes are removed from immunized mice and teased apart to prepare a single
cell suspension which are mixed with the myeloma cells derived from in cultures. It is important
to have healthy and actively dividing (i.e., log phase) myeloma cells on the day of the fusion.

A fusogenic agent is added to the mixed myeloma and spleen cells. Polyethylene glycol
(PEG) is the most common agent used for the fusion of cells. Other fusogenic agents will work,
though. The spleen cells and myeloma cells are mixed and co-pelleted by centrifugation. A
small volume of PEG solution in media is added and the cell pellet is gently dislodged. PEG
fuses the plasma membranes of adjacent cell resulting in a heterokaryon. PEG is toxic and
exposure needs to be minimized. The cells are washed to remove the PEG, resuspended in HAT
medium (see below), and then generally plated in 24-well plates.

SELECTION

Not all of the cells will fuse following PEG treatment and many of fusions will be
homotypic. The unfused myeloma cells will tend to overgrow the fused cells in culture, and
therefore, it is necessary to select for fused cells. The use of hypoxanthine-guanine phosphor-
ribosyl transferase (HGPRT) deficient myeloma cells is a commonly used genetic selection
scheme for selecting fused cells. HGPRT is a crucial enzyme in the purine salvage pathway.
Selection of fused cells is achieved by adding hypoxanthine, aminopterin and thymidine (HAT)
to the culture medium. Aminopterin is an inhibitor of dihydrofolate reductase (DHFR) which is
a crucial enzyme for both de novo purine and pyrimidine synthesis. Hypoxanthine and
thymidine are nucleotides utilized in the purine and pyrimidine salvage pathways, respectively.
HGPRT-deficient myeloma cells will die in the presence of aminopterin since they are and
cannot salvage purines and de novo purine synthesis is inhibited. Spleen cells contain an intact
HGPRT and can therefore salvage, but do not have the ability to grow in culture. Therefore,
Balb/c Mouse

MOPC 21 (tumor)

P3K (cell line)

P3-X63Ag8 (HGPRT-)

X63-Ag8.653 (IgG-)
125
only hybridomas (i.e., fusions between myelomas and spleen cells) are able to grow in the
presence of HAT medium. The myeloma cells provide the ability to grow in culture and the
spleen cells provide a functional HGPRT. The HAT medium is usually replaced with HT
medium after 7-10 days after the fusion to minimize the aminopterin toxicity. Hypoxanthine
and thymidine are still necessary, since aminopterin slowly decays and is still present in the
medium. The plates are monitored for the next 2-4 weeks for the appearance of colonies.



SCREENING

The goal of the screening step is to identify wells containing the mAb of desired
specificity. The wells may contain hybridomas that do not secrete antibody or secrete irrelevant
antibody. Media from wells exhibiting cell growth are removed and tested for antibody. The
immunoassay that is used for screening needs to be worked out before reaching this point. Cells
producing mAb of desired specificity are expanded and stabilates frozen.

CLONING

Wells will sometimes contain mixtures of hybridomas. In addition, some of hybridomas
will lose the ability to produce antibody due to chromosome loss or rearrangement. Cloning the
hybridoma of interest will prevent irrelevant hybridomas from overgrowing the hybridoma of
interest. Two common methods utilized for cloning hybridomas are soft agar and limiting
dilution. In soft-agar cloning the hybridomas are plated in a semi-solid medium such as a low
concentration of agar. Single cells will yield colonies that can be expanded. The advantage of
soft-agar cloning is that a colony originating from a single cell is selected. However, many
hybridomas will not grow in soft agar. In limiting dilution cells are plated in 96-well plates at a
calculated concentration of one cell per well. With both methods it is recommended to subject
126
the hybridoma to several rounds of cloning until a stable hybridoma is obtained.

PRODUCTION

Large quantities of mAbs can be produced by either in vitro culture or through the
generation of ascites tumors. mAb produced in vitro is sometimes too dilute and will generally
be contaminated with bovine serum components. Therefore, it may be necessary to purify
and/or concentrate the mAb. Ascites are produced by injecting the hybridoma into the
peritoneal cavity of a mouse. The hybridoma cells will produce an ascites tumor and secrete
mAb into the peritoneal cavity. The ascites fluid is then collected several weeks later. Ascites
can produce very high concentrations of antibody. The ascites fluid will be slightly contam-
inated with other mouse IgG.


127
APPENDIX. B-CELL DIFFERENTIATION

Lymphoid stem cells found in the bone marrow undergo gene rearrangements in the
variable regions of the Ig-heavy and the Ig-light chain genes. Cells undergoing heavy chain
rearrangements are called progenitor (pro) B-cells and those undergoing rearrangements of
the light chains are precursor (pre) B-cells. These rearranged genes will then be expressed as
IgD and IgM on the surface of the mature B-cells. The mature B-cells are a heterogeneous
population of cells which all express different rearrangements of the immunoglobulin (Ig)
genes. Each rearrangement results in an immunoglobulin with a different epitope specificity.

These naive B-cells (not yet exposed
to antigen) are found in the secondary lym-
phoid tissues where they will be exposed to
potential immunogens. Binding of antigen to
the surface immunoglobulin will stimulate
those particular cells to proliferate and differ-
entiate into plasma cells. The activation of
most B-lymphocytes will also require the
participation of activated T-helper cells. In
particular, Th
2
cells secreting IL-4, IL-5, IL-6
and IL-10 stimulate the proliferation and
differentiation of the B-cells. Bound antigen
is also taken up by receptor-mediated endo-
cytosis, processed and presented on the B-
cell surface. These B-cells function as anti-
gen presenting cells and stimulate T-cells.

The plasma cells no longer express
surface immunoglobulin and begin to secrete
IgM. The switch from membrane IgM to
secretory IgM involves an alternate mRNA splicing. Plasma cells can also undergo a class
switching resulting in the production of IgG (primarily in blood), IgA (primarily associated
with mucosa), or IgE (associated with allergenic cells). IL-2 and INF- are needed for class
switching. Immunoglobulin class switching can be accomplished at either the DNA level
(i.e., gene rearrangement) or at the mRNA level (i.e., differential splicing). An affinity
maturation due to mutations in the hypervariable region is also observed.

The disappearance of the antigen leads to apoptosis of many of the plasma cells, or a
conversion to memory cells. A second encounter with the same antigen leads to an
activation of these memory cells and a more rapid response and a higher steady-state level
of antibody. In addition, the secondary response is primarily IgG.

128
CHAPTER 15--IMMUNOASSAYS

Due to their specific and high affinity interactions
with epitopes, antibodies are powerful reagents for the
analysis of proteins or other antigens. Antibodies are
composed of conserved regions and variable regions. The
variable regions are responsible for recognizing epitopes
within antigens. In the case of proteins, the epitopes are
typically 4-6 amino acids and are usually specific for a
particular amino acid sequence. The antibody molecule is
made up of heavy chains and light chains which are
joined by disulfide bonds. The heavy chain defines the
type of immunoglobulin. For example, IgM and IgG
contain and heavy chains respectively. In addition,
different chains define subclasses of IgG. The light chains ( and ) are shared between the
different immunoglobulin classes and subclasses. IgG is composed of two functional domains
separable after digestion with papain: F
ab
and F
c
. The F
ab
fragment contains the epitope binding
domain or the antibody activity, and the F
c
domain is characterized by complement binding in
some IgG subclasses.

One possible application of antibodies is affinity chromatography. Antibodies are
covalently attached to a matrix such as agarose to make affinity columns. More often, though,
antibodies are used in various analytical assays. These include the enzyme-linked immuno-
sorbent assay (ELISA), the radioimmunosorbent assay (RIA), immunofluorescence (IFA),
immunoblotting (also called Western blotting) and immunoprecipitation. Immunoassays are
used to detect, quantify and characterize antigens. The assays can also be used to detect and
quantify antibodies.

The general procedure used in all immunoassays
is 1) to allow antibody-antigen complex formation, and 2)
detection of the antibody-antigen complex. Antigen-
antibody complexes are formed by incubating the antibody with an extract containing the
antigen. Alternatively, the antigen is absorbed or fixed to a solid support and then antibody
binds to the immobilized antigen. The detection of the complex will depend on the assay being
perform (Box). One method is to directly label the antibody (or antigen) with a detectable
marker (eg., radioactivity, fluorescence, or enzyme). Alternatively, a labeled secondary
antibody (prepared in another species) that recognizes the primary antibody can be used to
detect the complex. Similarly, either protein A or protein G labeled with a marker could be used
to detect the antigen-antibody complex. Protein A and protein G are proteins isolated from
Staphylococcus areus or Streptococcus, respectively, that bind to the F
c
-region of IgG. Protein
G recognizes more immunoglobulin subclasses, as well as a wider range of species, than protein
A.


directly label 1
o
antibody
anti-antibody (2
o
Ab)
protein A or protein G
129
IMMUNOPRECIPITATION

Immunoprecipitation involves the purification of antigen-antibody complexes, which
are then analyzed by gel electrophoresis. This method provides a means to identify and
characterize proteins in conjunction with gel electrophoresis. Initially, the technique involved
the formation of large supramolecular antigen-antibody complexes due to polyclonal sera that
recognized multiple epitopes on the protein. These large protein complexes are collected by
centrifugation. Formation of the large complex is dependent on high concentrations of both
antigen and antibody and the complexes will often contain contaminating proteins. The ability
to purify the antigen-antibody complex with bacterial proteins that bind immunoglobulin has
greatly increased the sensitivity and specificity of immunoprecipitation.

A typical first step in immunoprecipitation is
to radiolabel the protein(s). This is easily accomp-
lished in situations where in vitro cell culture is
possible by metabolically labeling the protein with
radioactive amino acids. Proteins can also be labeled
by chemical means such as iodination as long as the
treatment does not adversely affect the epitopes recognized by the antibodies. It is not necessary
to label the protein if it can be detected by other means such as protein staining or immuno-
blotting (see combined immunoprecipitation and immunoblotting below).

Proteins are then solubilized under conditions which do not affect the ability of the anti-
body to recognize the epitope. For example, some antibodies recognize 'conformational' epi-
topes and denaturing agents will destroy the epitope. Generally, non-ionic detergents (eg.,
Triton X-100) do not adversely affect the immunoprecipitation assay. It still may be possible to
use immunoprecipitation for the analysis of proteins which are only soluble under denaturing
conditions if a non-conformational dependent (i.e., linear) epitope is recognized by the anti-
body. However, the denaturing agent will have to be removed before adding the antibody since
the structure of the antibody must be maintained for epitope recognition. For example, proteins
can be solubilized with SDS followed by the addition of a 10-20 fold excess of Triton X-100 (or
another non-ionic detergent). The excess SDS will be incorporated into mixed micelles and
therefore will have little affect on the antibody when it is added. Enough antibody should be
added to precipitate all of the antigen in a sample, especially if quantitative interpretations are
going to be made.

In general, the formation of antigen-antibody complexes is rapid and incubations of 30-
60 minutes are sufficient for maximum recovery. Longer incubations may be necessary for low
concentrations of the target protein. Proteolysis can also be a problem during the incubation
step. Protease inhibitors should be added and the incubations carried out on ice to minimize
losses.

The isolation of the antigen-antibody complex is
accomplished by an affinity purification with protein A or
protein G conjugated to agarose beads. Historically, fixed
intact Staphylococcus areus bacteria were used to isolate Ag-
1. Label the Ag
2. Solubilize the Ag
3. Mix and incubate Ag and Ab
4. Isolate Ag-Ab complex
5. Electrophoresis/fluorography
agarose
G
130
Ab complexes. Protein A and protein G bind to the F
c
region of IgG with an high affinity. In
general, protein G binds with a higher affinity to a wider range of immunoglobulin subclasses as
well as recognizing immunoglobulins from more species than protein. These bacterial proteins
have a higher affinity for antibody with bound antigen and are therefore the protein A (or G)
agarose is usually added after the formation of the antigen-antibody complex. But all three
components (antigen extract, antibody, and affinity matrix) can be mixed simultaneously.
Generally, incubation for 30-60 minutes with continuous rocking is sufficient for complete
recovery of the antigen-antibody complexes.

The antigen-antibody complexes bound to the protein A (or G) are collected by centrifu-
gation and washed several times to remove contaminating proteins. Antigen-antibody com-
plexes are then eluted from affinity matrix by boiling the agarose beads in electrophoresis
sample buffer containing SDS and -mercaptoethanol. Bound proteins can also be eluted with
low pH and quickly neutralized if non-denatured protein is need for subsequent analyses. The
sample is then subjected to SDS gel electrophoresis and immunoprecipitated proteins are
detected by autoradiography or fluorography depending on the isotope. It may also be possible
to detect abundant antigens by protein staining. Alternatively, it
may also be possible to transfer the electrophoresed proteins to a
membrane and carry out immunoblotting (see below).

IgG heavy and light chains will also be contained in the
sample and can be observed as approximately 55 (IgG-H) and 25
(IgG-L) kDa bands (Figure). Most of the time this will not cause
any problems unless the target protein is of a similar size and the
immunoglobulin subunits interfere with the protein of interest.The
immunoglobulin bands are easily control for by including an
immunoprecipitation blank (B) which contains everything (i.e,
antibody and buffers) except the sample (S) being analyzed. Bands
appearing in the sample (arrow), but not the blank will be the
protein(s) recognized by the antibody.

Immunoprecipitation can be used to detect a protein in a biological sample and in con-
junction with SDS-PAGE to determine its molecular weight. Since the formation of the Ag-Ab
complexes can be carried out under non-denaturing conditions it is also possible to determine
the quarternary structure of a protein and to analyze proteins that bind to a target protein. Other
proteins binding to the protein recognized by the antibody will be co-immunoprecipitated with
the target protein if the protein-protein interactions are not disrupted by antibody binding (or
visa versa). Generally the information obtained by immunoprecipitation is qualitative. To obtain
reliable quantitative information is will be necessary to titrate the antibody and to run controls
demonstrating that all of the available target protein has been immunoprecipitated.

131
IMMUNOBLOTTING

Immunoblotting, also called Western blotting, is
another technique used in conjunction with gel electropho-
resis. In this case, proteins are first subjected to gel electro-
phoresis, transferred to a membrane support and the protein
of interest is detected with an antibody.

The first step in immunoblotting is antigen prepara-
tion and electrophoresis. SDS-PAGE is the most common electrophoretic system, but other gel
electrophorectic systems can also be used. After electrophoresis the separated proteins are
transferred to a membrane support such as nitrocellulose or PVDF (Immobilon-P). PVDF mem-
branes bind more protein and are generally superior to nitrocellulose. Before the transfer the gel
is usually equilibrated with the transfer buffer containing methanol to remove excess SDS.
Excessive SDS can inhibit the binding of proteins to the membrane. The transfer of proteins
from the gel to the membrane is carried out electrophoretically in a special apparatus. The gel is
positioned between two electrodes that generate an uniform electric field across the thickness of
the gel (Figure). A membrane is placed on the anode side of the gel. Filter papers are used to
maintain buffer contact with the gel. Proteins migrate out of the gel and a replica of the gel is
formed on the membrane. The completeness of the transfer can be assessed through the use of
prestained proteins as size markers or by staining the gel after the transfer.



Following the transfer, the membranes are washed with a buffer which removes any
loosely bound proteins and which will 'block' the non-specific binding of proteins (i.e.,
antibodies). Two common blocking agents are the detergent Tween-20 and a powdered-milk
solution, called BLOTTO in lab jargon. The detergent (or milk proteins) will bind any free sites
on the membrane that are not already bound with protein and prevent non-specific antibody
binding. The membrane is then incubated with the primary antibody (diluted with the blocking
buffer). The dilution and time of incubation will depend on antibody affinity and the amount of
1. Antigen electrophoresis
2. Transfer to membrane
3. Blocking
4. Incubate with 1
o
antibody
5. Wash
6. Incubate 2
o
antibody
7. Wash
8. Develop with substrate
132
the target protein on the membrane. Typically incubations are carried out for one hour at room
temperature. Alternatively, overnight incubations at 4
o
can be carried out.

The membranes are then washed extensively to remove any non-specifically bound
antibody followed by an incubation with the secondary antibody conjugated with an enzyme.
This secondary antibody recognizes immunoglobulin from the species of the primary antibody
and are purchased. The secondary antibody will bind the primary antibody which is bound to
the antigen. The membranes are then incubated with substrates of the enzyme which will form
insoluble colored products (see Table in section on ELISA for examples). This results in a
colored precipitate being formed on the membrane in the position of the target protein.
Chemiluminescent substrates can also be used instead of the precipitating chromogenic
substrates (see section on non-radioactive probes in chapter on nucleic acid blotting). Alterna-
tively,
125
I-Protein A in conjunction with autoradiography can be used to detect the Ab-Ag
complex.

Both immunoblotting and immunoprecipitation can identify proteins recognized by
an antibody in conjunction with SDS gel electrophoresis. The choice of the assay will be
affected by the experimental design, antigen concentration, antigen solubility and the ability
of the antibody to recognize the antigen in the two assays. As an example of experimental
design, immunoblotting will provide information about the total amount of antigen present,
whereas immunoprecipitation of metabolically labeled proteins can provide information
about the newly synthesized protein and/or its rate of synthesis. The limited amount of
protein which can be electrophoresed may preclude the use of immunoblotting for the
analysis of low abundance proteins, whereas immunoprecipitation can be used to
concentrate low abundance proteins from relatively large volumes.

Experimental Design
eg., synthesis (IP)
Ag concentration
IP better for low abundance
proteins
Ag solubility
blot for insoluble proteins
Ab recognition
conformational dependent
epitopes
4
o
structure

Another factor involved in the success of either immunoprecipitation or immuno-
blotting may be whether the antibody recognizes the protein of interest. In general, native
non-denatured proteins are recognized by the antibody in the immunoprecipitation assay,
whereas denatured proteins are recognized by the antibody in the immunoblotting assay.
The nature of the epitope will determine whether one, or both, assays will work. For
example, if the antibody recognizes a linear epitope that is exposed on the surface of the
protein, it will likely function in both assays. However, if the antibody recognizes a
conformation epitope, then only immunoprecipitation can be used. In the situation where an
antibody recognizes a linear epitope that is not exposed in the native protein it will be
133
necessary to denature the protein before carrying out the immunoprecipitation. The quarter-
nary structure of a protein may also influence the choice of the assay. For example, an
antibody that only recognizes one subunit of a multisubunit protein will only produce one
band if analyzed by immunoblotting, whereas all of the protein subunits may be detected in
the immunoprecipitation assay.

One problem with the immunoprecipitation assay is that it is not always possible to
radiolabel the antigen of interest, and therefore detecting the target protein is difficult. Immuno-
blotting provides a very sensitive method of detecting proteins after gel electrophoresis, but is
limited by the amount of protein which can be electrophoresesed. Affinity purification of the
target protein by immunoprecipitation and detecting the protein by immunoblotting possibly
overcomes these limitations. Following immunoprecipitation and gel electrophoresis, the
proteins are transferred to a membrane and analyzed by immunoblotting. This does require that
the antibody works in both assays, or that distinct antibodies for each assay are available. A
blank immunoprecipitation which contains the antibody, but no antigen-containing extract,
should also be prepared. This blank will serve to identify the IgG heavy and light chains which
may be detected with the secondary antibodies.


IMMUNOFLUORESCENCE

Immunofluorescence is used to localize antigens to
specific cell types or subcellular compartments. The assay
can also be adapted for the detection and quantification of
antibody. The general procedure is to incubate cells or tissue with the primary antibody and
detect the antigen-antibody complex with a secondary antibody conjugated with a fluoro-
chrome. Fluorescein and rhodamine, which fluoresce green and red, respectively, are two
common conjugates. The location of the protein recognized by the primary antibody is
determined by examining the sample under UV illumination (see Section of Fluorescence
Microscopy). It is also possible to use an enzyme-conjugated secondary antibody and develop
with substrates that form precipitates. The location of the precipitates is then detected by
conventional microscopy. This variation is not widely used, though.

The preparation of the cells or tissue for immunofluorescence is a crucial step that
depends on the type of cells being analyzed and upon the application. As for microscopy in
general, tissues need to be sectioned with a microtome and mounted onto microscope slides
before immunofluorescence analysis. Adherent tissue culture cells are usually grown on cover
slips or microscope slides. Cells grown in suspension can be smeared or dried onto microscope
slides or the antibody incubations and washings can be carried out with the cells in suspension.

The fixation procedure should not affect the ability of the antibody to recognize the
antigen. Organic solvents, such as ethanol, methanol and acetone, or paraformaldehyde are
routinely used. Glutaraldehyde is not a good fixative since the bonds formed by the crosslinking
of glutaraldehyde exhibit a strong endogenous fluorescence. Fixation also permeabilizes the cell
so that the antibody has access to internal cellular structures. The permeability can be enhanced
by treating with a detergent after the fixation. Intact and unfixed cells can be used to
1. Prepare cells or tissue
2. Incubate 1
o
antibody
3. Wash
4. Incubate 2
o
antibody
5. Wash
6. View under UV illumination
134
demonstrate the surface localization of an antigen.

The subcellular structures recognized by the antibodies will appear as a bright image
against a dark background. Comparing the epifluorescence with the bright field image is used to
localize the protein of interest to a particular subcellular compartment or to a particular cell type
within a tissue. Fluorescent dyes which bind to particular subcellular compartments can also be
used to assist in the localization of the protein of interest. Dual-labeling experiments can also be
used to demonstrate the extent of co-localization between two different proteins. The most
convenient method is to used antibodies generated in different species (eg., mice and rabbits) as
the primary antibody and then used species specific secondary antibodies conjugated with
different fluorochromes. If only antibodies raised in the same species are available then it will
be necessary to purify the two antibodies and conjugate them with different fluorochromes. The
two different fluorochromes can be viewed separately through the use of filters specific for each
fluorochrome. It is also possible to digitally merge the two images to demonstrate the extent of
co-localization between the two proteins.

IMMUNOGOLD ELECTRON MICROSCOPY

The resolution of immunofluorescence is essentially the same as that for light micro-
scopy. Immunogold electron microscopy can more precisely localize proteins. The principle of
immunogold electron microscopy is the same as immunofluorescence in that antibodies are
bound to the protein of interest and then the location of these antibodies are determined through
the use of secondary antibodies conjugated with colloidal gold. It is also possible to use to use
enzyme-conjugated secondary antibodies in combination with substrates that form an electron
dense precipitate. In practice, though, localizing antigens at the level of the electron microscope
is more difficult and subject to more problems.

Samples need to be fixed and embedded for electron microscopy (see Section on
Electron Microscopy). However, these fixation and embedding procedures are quite harsh and
typically destroy epitopes. Gentler fixation procedures are generally needed and embedding is
done in resins which do not require heating. In general, fixation conditions which result in good
preservation of ultrastructure tend to lead to a loss in the ability of the antibody to recognize the
antigen. Following fixation and embedding, the samples are sectioned and mounted onto grids.
The incubations with primary or secondary antibodies and washes are carried out by floating the
grids on drops of buffer. The secondary antibodies are conjugated with colloidal gold. Different
sizes of gold particles are available for dual labeling experiments. The gold particles are
electron dense and are visible when examined by electron microscopy.

Cryo-electron microscopy can be used to overcome some of the problems associated
with the loss of antigenicity during the preparation of samples for electron microscopy. Special
micromes and stages which keep the sample frozen in liquid nitrogen are used. This allows for
lower concentrations of glutaraldehyde to be used. However, the resolution is much less in
cryo-electron microscopy than conventional transmission electron microscopy.

The relatively large size of the gold particles (5-15 nm) poses some accessibility
problems due both to the nature of the fixed cells and the embedding polymers. Furthermore,
135
many antibody molecules bind to a single gold particle resulting
in a large complex (Figure). One method to overcome these
accessibility problems is to use secondary antibodies conjugated
with ultrasmall gold (<1 nm). These small gold particles are not
readily visible under the electron microscope and need to be
treated with silver which coats the gold particles (i.e., silver
enhancement). In addition, the ultrasmall gold conjugated
secondary antibodies are more similar to the fluorochrome conjugated secondary antibodies
since one gold particle will tend to bind one antibody molecule. This allows for pre-embedding
experiments in which the fixed and permeabilized cells are incubated with the primary and
seconday antibodies in suspension before the sample is embedded. Therefore it is possible to
carry out the assay under conditions simlar to that used for immunofluorescence and avoid the
problems associated with embedding.

ELISA (AND RIA)

The enzyme-linked immunosorbent assay (ELISA)
involves absorbing an antigen a solid support such as a
membrane (as used in immunoblotting) or a 96-well micro-
plate. Proteins which have not been separated by electrophore-
sis can be bound to membranes and analyzed with primary and
secondary antibodies as in the immunoblotting procedure.
Such analyses are often called dot blots. The more common format is to absorb the antigen to
the wells of a 96-well microplate and to use substrates that produce a colored soluble product.
The ELISA is particularly well suited for the analysis of large numbers of samples and most of
the procedure can be automated.

The first step is to bind the antigen to the plate. The antigen range from a purified
protein or peptide to a crude protein extract. It is also possible to use whole fixed cells. The
choice of antigen will depend on the application and the nature of the antibodies being tested.
Binding to 96-well microplates is accomplished by incubating the wells with a solution
containing the antigen. Protein binding is due to hydrophobic interactions between the protein
and the plastic. After antigen is bound to the plates, they are then blocked to eliminate non-
specific binding. Common blocking agents are bovine serum albumin (relatively cheap protein),
Tween-20 (detergent), or powdered milk (very cheap protein). The function of the blocking step
is to coat any protein-binding sites on the plate that could bind antibody non-specifically.

The antigen-containing wells are then incubated with the primary antibody. The proper
antibody dilution must be determined empirically. Serial dilutions are often carried out and the
minimum titer giving a positive response will be a quantitative measure of antibody concen-
tration. ELISAs are typically carried out in duplicate or triplicate. It is important to include
negative (without antigen) controls on the same plate. Positive controls can also be included in
situations where the antigen is being evaluated. After incubation with the primary antibody the
wells are washed extensively to remove any unbound antibody.

1. Antigen binding
2. Blocking
3. Incubate 1
o
Ab
4. Wash
5. Incubate 2
o
Ab
6. Wash
7. Develop with substrate
136
The wells are then incubated with a secon-
dary antibody that is conjugated with an enzyme.
Alkaline phosphatase and horse-radish peroxidase
are two common enzymes. These reagents are
obtained commercially and recommended dilutions
will be supplied by the manufacturer. It is possible
to buy secondary antibodies that are class or sub-
class specific (i.e., made against the heavy chain).
Second antibodies are made against the whole mole-
cule will usually recognize the light chains which
are shared between IgG and IgM. The plates are
again washed to remove any unbound antibody. The final step is to develop the plates with a
chromogenic soluble substrate (Table). The intensity of the color will be proportional to the
amount of secondary antibody which proportional to the amount of specific primary antibody in
the sample. Alternatively a constant amount of a defined antibody can be used to quantify the
amount of a specific protein in a sample. An ELISA reader is a spectrophotometer designed to
read the individual wells in a 96-well microplate. Many ELISA readers are interfaced with a
computer to assist in data management.

The RIA is essentially the same as the ELISA, except
125
I-protein A is used instead of
enzyme-conjugate secondary antibody for the detection of the Antigen-antibody complex. The
amount of radioactivity bound to the membrane filters or wells is determined instead of
developing with substrates. The RIA was developed first, but the ELISA is more popular since
radioactivity is not needed.

Variations to the ELISA can also
be carried out. For example, the plates can
be coated with the antibody instead of the
antigen. The antibody will then 'capture'
the antigen when extracts are added to the
wells. In this case the antigen will need to
be radioactive or have some type of
measurable activity. If two antibodies with different specificities are available it may be
possible to create a 'sandwich' type of assay. Monoclonal antibodies are especially useful for the
development of antigen capture assays.

DIRECT VS. INDIRECT

In summary, ELISA, immunoblotting
and IFA are all techniques in which antigen is
immobilized, incubated with primary antibody and the bound antibody is then detected. The
bound antibody is usually detected with a second antibody, which recognizes the first antibody
and is conjugated with a detectable marker (eg., radioactivity, enzyme, fluorescence). The use
of a secondary antibody is referred to as indirect detection.

Examples of ELISA Substrates
Substrates
Enzymes Soluble Precipitating
AP pNPP BCIP/NBT
HRP TMB DAB
AP (alkaline phosphatase); HRP (horse
radish peroxidase); pNPP (p-nitrophenyl
phosphate); BCIP (5-bromo-4-chloro-3-
indolyl phosphate); NBT (nitro blue tetra-
zolium); TMB (3,3',5,5'-tetramethyl-
benzidine); DAB (3,3-diaminobenzidine)
Direct vs. Indirect
less steps
less bkg (?)
dual label
convenience
amplification (?)
137
It is also possible to label the primary antibody
with the detectable marker and use the primary antibody
to directly detect antigen. Direct detection methods are
quicker to carry out since an antibody incubation step and
the associated washing steps are eliminated. In addition,
there is sometimes less background associated with direct
detection methods. However, the primary antibody must
be first purified and conjugated with the appropriate
label. The availability of secondary antibodies from
commercial sources tends to make the indirect methods
easier to carry out. In addition, some signal amplification
may occur during the secondary antibody step.

The most common detectable markers used to label antibodies are: radioactivity
fluorochromes, biotinylation and enzyme cross-linking. In all of these labeling procedures, it is
important that the modification does not affect the antibody's ability to recognize the antigen.
Iodination is the preferred method for radiolabeling antibodies. Many iodination methods are
gentle and do not adversely affect antibody structure. In addition, mAbs can be metabolically
labeled by incubating the hybridomas with radioactive amino acids and harvesting the mAb.
Arginine is a good amino acid for the labeling of mAb.

Fluorescent antibodies can be made with commercially available isothiocyanate
derivatives of fluorescein and rhodamine. These derivatives react with primary amines (i.e.,
primarily lysine). The efficiency of labeling can be conveniently determined by spectropho-
tometry. For example, the A
495
/A
280
ratio should be between 0.3-1.0 following conjugation with
fluorescein and the A
575
/A
280
ratio should be 0.3-0.7 following conjugation with rhodamine.
Biotinylation with a succimide ester of biotin also labels primary amines. The labeling of free
amines is usually mild and does not affect antibody function. However, the labeling should be
kept to a minimum.

Enzymes can also be crosslinked to antibodies. The three most common enzymes are
horse radish peroxidase, alkaline phosphatase and -galactosidase. Methods of cross-linking
include glutaraldehyde, periodate treatment and bifunctional cross-linking agents.

Biotin-Avidin Detection Systems. Biotin-avidin based
detection systems are also available. Biotin-conjugated secondary anti-
body or protein A is used to bind the primary antibody. This is
followed by an incubation with avidin (or streptavidin) conjugated
with detectable marker (i.e., fluorescein, enzyme,
125
I). One advantage
of this detection system is that the same secondary antibody can be
used in different immunoassays, thus reducing the need to purchase
many different secondary antibodies. However, one still needs the
appropriately labeled avidin for all of the different assays. Avidin has a
high affinity for biotin and several biotin molecules can be incorporated into a single antibody
molecule; thus increasing the sensitivity of the assay. However, the use of the biotin-avidin
systems requires an additional incubation step and the associated wash steps.
Radiolabeling
iodination
metabolically (Mabs)
Fluorochromes
fluorescein
rhodamine
Biotinylation
Enzyme Crosslinking
alkaline phosphatase
horseradish peroxidase
138

OTHER METHODS

Some older immunological methods that are still occasionally use are Ouchterlony,
immunoelectrophoresis and crossed immunoelectrophoresis. All of these methods depend on
the formation of large supramolecular complexes of antigen and antibody (as discussed in the
section on immunoprecipitation). In all of these methods the large antigen-antibody complexes
will precipitate within a gel. These methods are insensitive and depend on high antigen and
antibody concentrations, as well as a polyclonal sera. In Ouchterlony the antigen and antibody
are placed in separate wells and diffuse towards each other. Antigen and antibody complexes
will form within the gel at a position between the two wells and precipitate. Immunoelectro-
phoresis is a similar method, except that the antigen sample is first subjected to electrophoresis
to separate proteins and then the antibody is allowed to diffuse into the gel. In crossed
immunoelectrophoresis the sample is electrophoresed in a second dimensional gel containing
the antibody. In both of the electrophorectic methods, the precipitates will form in association
with the specific proteins recognized by the antisera.

Comparison of Immunoassays
TECHNIQUE GENERAL PROCEDURE TYPICAL APPLICATIONS
ELISA
adsorb Ag to solid support
incubate with Ab
detect bound Ab
quantify Ab
quantify Ag
process large # of samples
BLOTTING
SDS-PAGE and transfer
incubate with Ab
detect bound Ab
identify subunit MW
quantify Ag
quantify Ab?
IFA
fix cells on slide
incubate with Ab
detect bound Ab
subcellular localization
quantify Ab
quantify Ag?


139






PART IV



Nucleic Acids and Recombinant DNA



Topics covered:

Nucleic Acid Structure and Isolation
Electrophoresis and Blotting
Polymerase Chain Reaction (PCR)
DNA Sequencing and Bioinfomatics
Recombinant DNA
141
CHAPTER 16--NUCLEIC ACID STRUCTURE AND ISOLATION

Nucleic acids encode information relating to cell structure and function. Cells have the
ability to make exact copies of their DNA and pass this information to daughter cells. DNA also
serves as a template for the synthesis of RNA, or transcription. The different types of RNA are
known as ribosomal RNA (rRNA), transfer RNA (tRNA), and messenger RNA (mRNA). The
mRNA is processed and translated into proteins. rRNA and tRNA are necessary components
for protein translation. Manipulating and sequencing nucleic acids are often an easier approach
to studying protein function and structure than isolating and characterizing proteins.

Nucleic acids are polymers of nucleotides. Nucleotides are composed of ribose (a
5-carbon sugar) and either a purine or a pyrimidine base at the 1'-position (Figure). The purine
bases are adenine (A) and guanine (G) and the pyrimidine bases are cytosine (C), thymine (T)
and uracil (U). Uracil is only found in RNA and thymine is only found in DNA. The
5'-hydroxyl of the ribose group can be phosphorylated. Unphosphorylated forms are called
nucleosides and phosphorylated forms are called nucleotides (Table). Nucleotides exist as
monophosphates, diphosphates or triphosphates and are designated by the nucleoside name and
the number of phosphates (eg., adenosine monophosphate, or AMP; adenosine diphosphate, or
ADP; etc). Nucleotides that make up DNA lack the 2' hydroxyl (hence the name deoxy-
ribonucleic acid) and are indicated with a lower case 'd'.

Nucleotide Nomenclature
Base Nucleoside Nucleotide
Adenine Adenosine AMP
Guanine Guanosine GMP
Cytosine Cytidine CMP
Thymine Thymidine dTMP
Uracil Uridine UMP



Oligonucleotides are formed via a phosphodiester bond joining the 5-phosphate from
one nucleotide with the 3-hydroxyl on another nucleotide (Figure). These phosphodiester
bonds form the phosphate backbone of the oligonucleotide. In addition, oligonucleotides have
a polarity with a phosphate group at the 5'-end and a hydroxyl group at the 3'-end. During DNA
and RNA synthesis nucleotides are added to the 3-hydroxyl by polymerases using nucleotide
triphosphates (NTPs) as the substrates. The nucleotide is cleaved between the first and second
phosphate resulting in the -phosphate being incorporated into the phosphate backbone.

DNA can also form a double-stranded molecule and the two strands are held together by
hydrogen bonds formed between A and T residues and between C and G residues (Figure). The
two strands of DNA are oriented in opposite directions in terms of the 5'-to-3' polarity. The
restricted base pairing between nucleotides and the opposite polarities of the strands result in the
two strands being complementary. During DNA replication and RNA transcription one strand
will serve as template for the synthesis of the complementary strand. The strands of DNA also
twist around each other to form a helix. RNA also forms secondary structures that result from
base-pairing between complementary nucleotides within a same strand. Similarly, small
142
circular DNA molecules can also form higher ordered structures known as super coils. DNA
free of proteins, or naked DNA, probably does not exists in cells. Instead, DNA is associated
with various DNA-binding proteins which package the DNA into chromosomes.

Nucleic Acid Structure



ISOLATION OF NUCLEIC ACIDS

Three major types of techniques, or combina-
tions of them, are employed in the isolation of nucleic
acids: differential solubility, absorption methods, or
density gradient centrifugation. The choice of method
will depend on the type of DNA being isolated (Box) and the application. A major goal of
nucleic acid isolation is the removal of proteins. The separation of nucleic acids from proteins is
generally easily accomplished due to their different chemical properties. In particular, the highly
charged phosphate backbone makes the nucleic acids rather hydrophilic as compared to proteins
which are more hydrophobic. Separating the different types of nucleic acids can be more
problematic in that they all have similar chemistries. On the other hand, though, this similar
chemistry results in a few basic procedures which are common to many nucleic acid isolation
protocols. Most nucleic acid isolation protocols involve a cell lysis step, enzymatic treatments,
differential solubility (eg., phenol extraction or absorption to a solid support), and precipitation.
Genomic (chromosomal)
Organellar (satellite)
Phage/Viral (ds or ss)
Plasmid (extrachromosomal)
Complementary (mRNA)
143

Cell Lysis. Nucleic acids must be solubilized from cells
or other biological material. This solubilization is usually carried
out under denaturing conditions such as: SDS, alkali, boiling or
chaotropic agents. These denaturing conditions efficiently
solubilize the nucleic acids and generally do not adversely affect
them. In addition, the denaturing conditions promote the removal of proteins during the
subsequent steps and inhibit the activity of nucleases which will degrade the nucleic acids.

Enzymatic Treatment. Another approach in the isolation of nucleic acids is to degrade
unwanted components. For example, inclusion of proteases (eg., proteinase K) in the lysate will
promote the removal of proteins. Proteinase K is still active at 55
o
in the presence of 0.5% SDS.
The elevated temperature and SDS improve solubility and inhibit any DNAse activity that may
be present in the lysate. Nucleases can also be used to remove unwanted nucleic acids. For
example, many DNA extraction protocols include a RNase treatment step, and visa versa. It is
important that the RNase be free of DNase activity. DNase-free RNase is easily prepared by
boiling commercial RNase for 10 minutes. The stability of RNase makes the preparation of
RNase-free DNase more difficult. RNase-free DNase should be purchased from a reliable
vendor or tested before it use.

Phenol Extraction. Phenol is an organic solvent that is used to separate
proteins from nucleic acids. Proteins are hydrophobic and partition in the
organic phase. Nucleic acids are highly charged and partition in the aqueous
phase. The advantages are of phenol extraction are that it is easy to carry out
and can be adapted to many applications. It is also easily applied over a wide
range of volumes (40 l to several ml). In particular, phenol extraction is
widely used for the isolation of high molecular weight genomic DNA (see
below).

Phenol extraction is accomplished by mixing the sample with an equal
volume of phenol which has been previously saturated with a Tris buffer at pH
8 containing EDTA and NaCl. The phenol should be molecular biology grade
phenol should and store at -20
o
until preparing the saturated solution. The
saturated solution is stored at 4
o
. Phenol is easily oxidized, as evidenced by
yellowing, and the oxidation products can break DNA. Oxidized phenol
should be discarded.

Depending on the application, the two phases are completely mixed by
vortexing, or gently mixed (eg., high molecular weight DNA). The phases are
separated by centrifugation and the upper aqueous phase, which contains the nucleic acids, is
retained. Proteins will often be visible as flocculent material at the top of the phenol phase. The
two phases need to be carefully separated in that the nucleic acids and proteins tend to be at the
interface. Leaving too much of the aqueous layer behind will lead to undue loss of material and
aspirating too close to the interface can include protein. The aqueous phase can be re-extracted
with phenol to remove more protein. Phenol is a hazardous waste material that needs to be
disposed of properly.
Cell Lysis
Enzyme Treatment
Phenol Extraction or
Adsorption Methods
EtOH Precipitation
144

A common variation of phenol extraction is a mixture of phenol:chloroform: isoamyl
alcohol (25:24:1). The more organic chloroform removes lipids, denatures more protein and
mixes less with the aqueous phase leading to more efficient extraction.

Ethanol Precipitation. Nucleic acids can be precipitated from dilute solutions with
ethanol. This precipitation can be a concentration step or a means to change buffers, especially
after phenol extraction. Typically either sodium acetate or potassium acetate, pH 5.0-5.5, is
added to a final concentration of approximately 0.3 M. The sodium and acidic pH will
neutralize the highly charged phosphate backbone and to promote hydrophobic interactions.
Two-to-two and a half volumes of ethanol are added and the sample is incubated as -20
o
. If the
nucleic acids are small in size and/or in low concentrations an extended incubation (several
hours to overnight) is needed. The precipitated DNA is collected by centrifugation. The pellet is
rinsed with 70% ethanol to remove any excess salt, dried and dissolved in the appropriate
buffer. A variation is to substitute ammonium acetate if the 'hard' salts are a problem. Another
modification is to use an equal volume of isopropanol (instead of 2-2 volumes of ethanol)
which minimizes the increase in sample volume.

Isolation of High Molecular Weight Genomic DNA. High
molecular weight chromosomal DNA is usually isolated by multiple
rounds of phenol extraction and enzyme treatments as discussed above.
Shear forces, which can break long DNA molecules, need to be avoided
during all steps and samples should never be vortexed. Therefore, the
phenol extraction is carried with gentle rocking for several hours. These
precautions against shear forces are not necessary in the isolation of low
molecular weight DNA. Another common modification at the ethanol
precipitation step is spool out the high molecular weight genomic DNA
on the end of a sealed Pastuer pipet. The precipitated DNA is wrapped
around the end of the pipet is then allowed to partially dry and then
dissolved in the appropriate buffer. This minimizes the contamination with
RNA and low molecular weight DNA fragments.

Plasmid Minipreps and Adsorption Methods. Historically, phenol
extractions were used for the isolation of most forms of nucleic acids. It is
now more common to use techniques based upon adsorption chromato-
graphy for the isolation of smaller DNA molecules, such as plasmids.
Various kits are available for the rapid isolation of small quantities of plasmid DNA. The
procedure consists of solubilizing the bacteria in an alkali solution followed by neutralization
with sodium acetate. The neutralization results in the precipitation of some of the protein and
the genomic DNA which is removed by centrifugation. The soluble material is then mixed with
a resin in the presence of chaotropic agents (usually guanidine hydrochloride). The resins are
usually either based on silica or diatomaceous earth. Under these conditions DNA binds to the
matrix, but proteins and RNA do not. The DNA is eluted in a low salt buffer. These methods are
rapid and yield a highly purified plasmid DNA which can generally be used directly in most
applications without further processing.

145
Another common application for an adsorption method is the isolation of DNA
fragments following gel electrophoresis. In this case the agarose gel piece containing the DNA
is dissolved in NaI, a chaotropic salt, and the DNA adsorbed to silica. The DNA is then eluted
with a low salt buffer and sometimes gentle heating.

ISOLATION OF RNA

Most RNA isolation protocols also
involve phenol extractions and are similar to
DNA isolations. However, there are some differ-
ences and special considerations (Box). In
particular, precautions against RNase activity
must be taken. RNase is an extremely stable and
active enzyme. Gloves should be worn at all times and sterile plasticware should be used
whenever possible to avoid introducing exogenous RNase to the sample. Glassware needs to
treated with DEPC-water and autoclaved to inactivate any RNase. Buffers should be prepared
from DEPC-water or RNase inhibitors included.

The cell lysis and solubilization of RNA will typically be carried out in guanidine salts
(especially guanidine thiocyanate). Guanidine is a strong chaotropic agent and will inhibit
RNases. The strong denaturing effect of this salt will also promote better phenol extraction. The
phenol extraction is the same as the DNA isolation except that the phenol is usually saturated
with a buffer of pH 5-6. The lower pH will result in some DNA partitioning in the organic
phase. If DNA contamination is a problem, it is possible to purchase RNase-free DNase. Long-
term storage is best as a precipitate in 70% ethanol at 20
o
.

Precipitation of RNA with LiCl. LiCl has been used to selectively precipitate RNA.
Large RNAs (rRNA, mRNA) are insoluble at high ionic strength, whereas small RNAs (tRNA
and 5S rRNA) and DNA generally remain soluble. Following either phenol or guanidine
extraction, an equal volume of 8 M LiCl is added. The sample is mixed vigorously and
incubated at -20
o
. The precipitate is collected by centrifugation and reprecipitated if necessary.

Affinity chromatography. Most eukaryotic
mRNA contains a stretch of A residues at its 3' end
which added post-transcriptionally. It is possible to
isolate mRNA by affinity chromatography on oligo-
dT columns (Figure). The RNA solution is passed
over an oligo-dT column under conditions which
promote base pairing. Only RNA with a polyA tail binds. The polyA RNA is eluted under
conditions (usually low salt and high temperature) which breaks the base pairing.


Oligo-dT Column
RNase inhibitors!
extraction in guanidine salts
phenol extractions at pH 5-6
treatment with RNase-free DNase
precipitation with LiCl
oligo-dT column
146
DENSITY GRADIENT CENTRIFUGATION

Density gradient centrifugation can also be used in
the analysis and isolation of nucleic acids. Double-stranded
DNA, single-stranded DNA, RNA and proteins all have
different densities (Box) and therefore can be separated by
isopycnic (i.e., equilibrium) centrifugation. CsCl is the standard medium for the density gradient
centrifugation of nucleic acids and are especially useful for the purification of large amounts of
highly purified DNA. The gradients are carried out in the presence of ethidium bromide which
fluoresces when bound to DNA. The DNA bands are detected by illumination with ultraviolet
light and easily recovered with a syringe and needle by puncturing the wall of the disposable
tube and aspirating the DNA. The CsCl can be removed by dialysis or by precipitating the
DNA.

The %G:C content affects the density of DNA (Figure). This can result in multiple
bands on CsCl gradients if DNA
composed different G:C compositions are
centrifuged. For example, minor bands,
called satellite DNA, are often observed
when total DNA from an organism is
analyzed by CsCl gradients. These satellite
bands are usually due to highly repetitive
DNA or organellar DNA. Mitochondrial
DNA is noted for its high A:T content.
These satellite DNA bands can be purified
from the genomic DNA by density
gradient centrifugation.

Nucleic acids can also be separated according to size by rate zonal centrifugation on
sucrose gradients. However, this method is not widely used since gel electrophoresis is
generally a more convenient method for the size fractionation of nucleic acids.

ANALYSIS AND QUANTIFICATION

The quality and quantity of isolated
nucleic acids can be determined spectrophoto-
metrically (Box). Nucleic acids have an A
max
of
260 nm and proteins have A
max
of 280 nm. The
A
260
/A
280
ratio is therefore indicative of the degree of purity of the nucleic acid. A
260
/A
280
ratios
of 1.6-1.8 or 1.8-2.0 are usually acceptable for DNA and RNA, respectively. The standard
extinction coefficient used for ssDNA, dsDNA and RNA are 0.03 ml/g, 0.02 ml/g and 0.025
ml/g, respectively. Formulas which take into account protein and other contaminants are also
available. Indirect spectrophometric assays for DNA quantification are also available, but rarely
used. In some instances fluorometry using fluorescent dyes that bind DNA and/or RNA is used
to determine nucleic acid concentrations.
Density in CsCl
DNA ~ 1.7 g/cm
3

Protein ~ 1.3 g/cm
3

RNA > DNA
ssDNA > dsDNA
A
260
1.0 50 g/ml
DNA
A
260
/A
280
1.6 - 1.8
A
260
1.0 40 g/ml
RNA
A
260
/A
280
~2.0
1.68
1.70
1.72
1.74
20 40 60 80
% GC base pairs
d
e
n
s
i
t
y

(
g
/
c
m
3
)
147
APPENDIX 1. COMMON CONVERSIONS

Weight Conversions

1 g = 10
-6
g
1 ng = 10
-9
g
1 pg = 10
-12
g
1 fg = 10
-15
g

Spectrophotometric Conversions

1 A
260
unit of double-stranded DNA = 50 g/ml
1 A
260
unit of single-stranded DNA = 33 g/ml
1 A
260
unit of single-stranded RNA = 40 g/ml

DNA Molar Conversions

1 g of 1,000 bp DNA = 1.52 pmole (3.03 pmoles of ends)
1 pmole of 1,000 bp DNA = 0.66 g

Protein Molar Conversions

100 pmoles of 100,000 dalton protein = 10 g
100 pmoles of 50,000 dalton protein = 5 g
100 pmoles of 10,000 dalton protein = 1 g

Protein/DNA Conversions

1 kb of DNA = 333 amino acids of coding capacity = 3.7 x 10
4
dalton protein
10,000 dalton protein = 270 bp DNA
50,000 dalton protein = 1.35 kb DNA
100,000 dalton protein = 2.7 kb DNA

Formulas

Picomole Ends per Microgram of Double-Stranded Linear DNA:
(2x10
6
)/(660xNumber of Bases)=pmole ends/ g

Exact Molecular Weight of an Oligonucleotide:
[(Ax312.2)+(Gx328.2)+(Cx288.2)+(Tx303.2)-61.0] = Molecular Weight of specific
oligonucleotide
149
CHAPTER 17--MODIFICATION OF NUCLEIC ACIDS

All DNA molecules are similar in terms of their biochemical and physical properties.
Therefore, unlike proteins, DNA techniques are not highly dependent upon the particular gene
being studied. However, genomic DNA is an extremely large molecule. For example, the
human genome contains approximately 2 x 10
9
base pairs (bp). The size of a gene for a 50 kDa
protein might be as small as 2000 bp (or 2 kb). Therefore, it can be quite difficult to identify and
characterize specific genes from an organism.

The study of specific genes involves manipulating nucleic acids. In particular, it is
possible to break DNA into smaller fragments, identify fragments of interest and to amplify
these fragments so that they can be analyzed. Enzymes are used to carry out these
manipulations of DNA and RNA. DNA modifying enzymes include: polymerases, ligases and
nucleases. Polymerases synthesize nucleic acids in a template mediated fashion. Ligases join
fragments of DNA. Nucleases cleave the phosphodiester bond between nucleotides. A wide
range of specificities are exhibited by nucleases (Box). Exonucleases remove nucleotides one at
a time from either the 5' or 3' end. In addition, some exonucleases exhibit substrate specificities
in terms of preferences for double-stranded (ds) or single-stranded (ss) DNA or RNA.
Endonucleases cleave the phosphodiester
bond in the middle of a oligonucleotide
and produce fragments. Such exo- and
endonucleases do have some applications
primarily directed at removing unwanted
types of nucleic acids or removing single
stranded overhangs from dsDNA.

RESTRICTION ENDONUCLEASES

In the late 1960's endonucleases that cleave DNA at specific sites were described. These
are called restriction endonucleases, or restriction enzymes. Restriction enzymes are found in
prokaryotes and they function to protect bacteria from phage (viruses of bacteria) infection. For
every restriction enzyme, there is a corresponding site-specific DNA modifying enzyme
(usually a methylase). Bacteria modify their own DNA by site-specific methylation which will
make the DNA resistant to digestion with that particular restriction endonuclease. Foreign DNA
from a bacteriophage will not be methylated at the appropriate restriction sites and thus be
degraded by the restriction endonuclease within the host bacteria. The name restriction enzyme
refers to restriction of some phages to
particular bacteria strains.

Three distinct classes of restriction
endonucleases, designated types I, II and III,
have been described (Box). Class II restric-
tion enzymes are the most useful since they
cleave DNA at the recognition site. Restric-
tion enzymes are named according to the
bacteria species from which they are isolated. The first letter is the first letter of the genus and is
Examples of Nucleases:
Bal 31 3'-exonuclease of ds or ssDNA
S1 ssDNA or ssRNA 5'-exonuclease
Mung Bean ssDNA 5'-exonuclease
DNase I ss or dsDNA (5' of pyr)
ExoIII 3'-exonuclease of dsDNA
RNase A ssRNA endonuclease (3' of pyr)
RNase T1 ssRNA endonuclease (3' of G)
Restriction Enzymes
Type I
cleavage occurs 400-7000
bp from recognition site
Type II
cleavage occurs adjacent or
within recognition site
Type III
cleavage occurs 25-27 bp
from recognition site
150
capitalized. The next two letters are the first two letter of the species and are in lower case.
Following these three letters (which are italicized) is an optional indicator for the strain of
bacteria. The last position is a Roman numeral which designates the number of restriction
enzymes discovered in that species and the order of their discovery.

The recognition sequences of type II restriction enzymes are generally 4-8 consecutive
nucleotides. Some restriction enzymes do allow for some degeneracy at a particular nucleotide.
This degeneracy can be restricted to a purines (R) or pyrimidines (Y), or be completely
unrestricted (N). Most recognition sites are palindromes in that both strands exhibit the same
sequence. The sequence complementarity and opposite orientations of the two strands leads to a
dyad symmetry. Hydrolysis of the phosphodiester bond results in the phosphate on the 5'-carbon
and the hydroxyl on the 3'-carbon. Both DNA strands are cut between the same two residues.
This will result in blunt ends if the cleavage site is in the exact center of the recognition site
(Figure). Alternatively, 5' overhangs (or extensions) or 3' overhangs of varying length will be
produced if the cleavage site is not in the center of the recognition site.
Isoschizomers are restriction enzymes isolated from different sources that recognize the
same sequence, but cleave at different positions resulting in different overhangs. In addition,
different recognition sites can produce identical overhangs (called compatible ends).
Reaction conditions. Restriction enzymes reproducibly cleave DNA at specific sites.
This feature provides a convenient way to break large DNA molecules into smaller
fragments for subsequent analysis and manipulation. Restriction digests are carried out by
incubating the DNA with the restriction enzyme. Each restriction enzyme has its own optimal
reaction conditions (pH, ionic strength, temperature, etc.). Manufacturers usually supply an
optimal 10X stock buffer with the enzyme. Appropriate amounts of DNA, the stock buffer,
other additives, the enzyme and water are mixed in a single tube. The DNA needs to be quite
Examples of
Restriction Sites
Enzyme Site
ClaI
A TC G A T
T A G CT A
EcoRI
GA A T T C
C T T A AG
FnuAI
GA N T C
C T N AG
HaeIII
G GC C
C CG G
HindII
G T YR A C
C A RY T G
HindIII
AA G C T T
T T C G AA
PstI
C T G C AG
GA C G T C
Blunt End (Sma I)


- CCCGGG- - CCC GGG-
| | | | | | | | | + | | |
- GGGCCC- - GGG CCC-


5' Overhang (EcoR I)


- GAATTC- - G AATTC-
| | | | | | | + |
- CTTAAG- - CTTAA G-


3' Overhang (Pst I)


- CTGCAG- - CTGCA G-
| | | | | | | + |
- GACGTC- - G ACGTC-

151
pure in that impurities in the DNA preparation (eg., salts, solvents, contaminating proteins, and
RNA) can inhibit the digestion. The incubation time will depend on the DNA concentration and
the amount of enzyme.

DNA can also be digested with mixtures of restriction enzymes. If the two enzymes
have the similar optimal reaction conditions, they can be mixed together and the reaction carried
out simultaneously. If not, the digestions are carried out sequentially by diluting the sample in
the appropriate buffer or by adding the required components. If necessary, the DNA can be re-
isolated after digestion with the first enzyme. The typically procedure is to phenol extract the
digestion mixture, precipitate the aqueous phase with ethanol, and dissolve the precipitated
DNA in the appropriate buffer for the second digestion.

Some restriction enzymes cleave
DNA with less specificity under certain
reaction conditions (Box). This is not the
same as the degeneracy in the recognition
site exhibited by some enzymes. For
example, at high enzyme or glycerol concen-
trations EcoRI will cleave at sites in which
only 5 of the 6 (GAATTC) nucleotides match
and this is referred to as star activity. Star activity is generally undesirable since it is random
and not reproducible. However, in some applications, digestions are carried under conditions
which promote star activity to introduce random cleavages.

Frequency of cutting. Theoretically restriction sites will be random within any genome.
The number of times a particular restriction enzyme will cut a genomic DNA can be
approximated from the size of the genome and its base composition. For example the
probability of finding any 6-base recognition sequence in a genome with 50% GC content is:

()()()()()() = 1/4096

This means that one would expect on average one restriction site every 4000 bp. These
restriction sites will not be evenly spaced every 4000 base pairs, but will occur randomly. For
example, the genome of E. coli (4 x 10
6
bp) will be digested into approximately 1000 fragments
of various sizes. The non-randomness of certain di- and trinucleotides can influence the actual
cutting frequencies.
Conditions Promoting Star Activity
high glycerol (>5%) concentration
high enzyme/DNA ratio (100 units/g)
low ionic strength (<25 mM)
high pH (>8)
organic solvents
substitution of Mg
2+

153
CHAPTER 18--ELECTROPHORESIS OF NUCLEIC ACIDS

Gel electrophoresis is commonly used to separate and analyze the nucleic acids. The
negatively charged phosphate backbone of nucleic acids gives them a uniform charge-to-mass
ratio. Therefore migration in gels is inversely proportional to the size of the nucleic acid (as
discussed for SDS gel electrophoresis of proteins). The structure of the nucleic acid (eg., linear,
circular, double-stranded, or single-stranded) will also affect migration in gels. It is possible to
distinguish DNA fragments that differ by as little as 1% and in some applications (eg., DNA
sequencing) fragments that differ in size by a single nucleotide can be resolved.

A wide range of sizes (<10 bp to
>20 kb) can be analyzed by gel electro-
phoresis. Agarose is the most common
matrix for the electrophoresis of nucleic
acids. Nucleic acids ranging in size from
100 bp to 20 kb can be resolved depending
on the agarose concentration (Table).
Acrylamide is used for smaller nucleic
acids and oligonucleotides.

Acrylamide gel electrophoresis can be carried out in the same apparatus used for protein
electrophoresis (i.e., vertical). Agarose gels are usually poured and electrophoresed in a
horizontal apparatus. The horizontal apparatus provides more support for the gel and are easier
to prepare. The appropriate amount of agarose is dissolve in the electrophoresis buffer by
heating the solution, which is then poured into a mold and allowed to cool. A well-forming
comb is inserted before the agarose polymerizes. The gel is submerged in the electrophoresis
tank buffer, the samples are loaded, and the appropriate voltage applied. Tracking dyes are used
to monitor the electrophoresis.



Staining with ethidium bromide (or other fluorescent dyes) is the most common method
for detecting nucleic acids after electrophoresis. The ethidium bromide can by co-polymerized
with the agarose gel. As the nucleic acids migrate through the gel the ethidium binds and
becomes fluorescent. Illumination with UV light will reveal the fluorescent bands. The progress
can be monitored during electrophoresis until the desire resolution is achieved. Alternatively the
gel can be stained post-electrophoresis by incubating in an ethidium bromide solution and
destaining with water. As little as 5-10 ng of DNA can be detected. Specific nucleic acids are
detected by transferring the nucleic acids to a membrane and hybridizing with a DNA probe
%
agarose
range
(kb)
%
acrylamide
range
(bp)
0.7 0.8-20 3.5 100-1000
0.9 0.5-7 5.0 80-500
1.2 0.4-6 8.0 60-400
1.5 0.2-4 12.0 40-200
2.0 0.1-3 20.0 10-100
NuSieve

3:1 agarose (FMC) is specially designed

to separate small DNA fragments (30-2000 bp).
154
(discussed below).

Shape also effects the migration of nucleic acids
in gels. For example, RNA and ssDNA can form
secondary structures due to internal base pairing. These
secondary structures will lead to a lower mobility.
Secondary structures are eliminated by running the gels
under denaturing conditions which prevent the
formation of base pairs. Chaotropic agents, such as urea,
formamide, or formaldehyde, are included in the gel
buffers for the electrophoresis of ssDNA and RNA if
sizes need to be accurately determined.

DNA can also exist as a circular molecule (eg.,
plasmids). Furthermore, the circular DNA can exist in
multiple conformations. In its natural state, circular
DNA is supercoiled due to additional superhelical
twists. Breaking a single bond in the phosphate
backbone of one strand will result in a relaxed circular DNA molecule. Breaking both strands in
the same position (eg., endonuclease) will result in the formation of a linear molecule.
Supercoiled circular DNA, relaxed circular DNA and linear DNA of the same molecular weight
will migrate at different rates through the gel. The typical order of migration from fastest to
slowest is supercoiled DNA, linear DNA, and relaxed circular DNA. However, the exact
migration is influenced by the agarose concentration, the applied current, ionic strength and
number of superhelical twists. Furthermore, circular DNA can form concatenated forms
(dimers, trimers, etc.). Therefore, multiple bands are often
observed in highly purified plasmid preparations. Complete
digestion with a restriction endonuclease which only cuts
once within the plasmid will result in the formation of a
single band corresponding to the linear DNA molecule.




Recovery of DNA from gels. DNA fragments are relatively easy to recover from gels
(Box). One possibility is to transfer the nucleic acids to membranes and analyze the DNA or
RNA by blotting techniques (discussed in Chapter 19). DNA can also be recovered by
electroelution. The regions containing the DNA fragment of interest is excised from the gel and
put into a dialysis bag with electrophoresis buffer. The dialysis bag is placed into an
155
electrophoresis chamber. The DNA fragment
will migrate out of the agarose but remain in the
dialysis bag. The DNA is then recovered by
precipitation.

Another method involves dissolving the excised agarose gel piece with NaI. A silica
resin which binds DNA is then added to the dissolve gel solution. Following several washes, the
DNA is eluted from the resin at low ionic strength (see section on purification by absorption
methods). Low-melting temperature (40-50
o
) agarose is also available. The DNA fragment is
excised and the gel is melted. Temperatures below 60
o
will rarely have an adverse affect on the
DNA. The fragment is then recovered from the solution by ethanol precipitation.

PULSE FIELD GEL ELECTROPHORESIS

Conventional gel electrophoresis has an upper limit to sizes that can be resolved. DNA
fragments larger than 20 kb tend to "snake" through the gel at the same rate. This phenomenon
is known as reptation. Pulsed field gel electrophoresis (PFGE) overcomes the problems of
resolution of large DNA fragments by using two sets of electrodes that are at fixed angles to
each other (see http://www.nalusda.gov/pgdic/Probe/v2n3/puls.html for review). The current is
alternated between these sets of electrodes at defined intervals. Separation is based upon the
ability of the smaller DNA molecules to reorient faster than larger DNA molecules. DNA
molecules up to 10 Mb can be resolved by PFGE.



Many lower eukaryotes have chromosomes in 0.5-3 Mb size range. PFGE provides a
means to characterize and map genes to the chromosomes of these organisms. In higher
eukaryotes PFGE is used for long-range restriction mapping with restriction enzymes that cut
infrequently.

After the initial description of PFGE in 1984, several variations of the method were
developed (Box). All are based on changing the direction of migration and the differences
between the methods are primarily in the positions of the electrodes. The contoured-clamp
homogeneous electric field, or CHEF, is probably the more popular method. The electrodes in
the CHEF apparatus are arranged in a square or hexagonal contour (Figure) leading to the
generation of a homogeneous electric field. This results in the lanes being straight and allows
for an easy comparison of different samples. In many of the other variations, the lanes run at an
angle and are difficult to compare.

transfer to membranes
electroelution into dialysis bags
NaI/glassmilk (eg., GeneClean

)
low-melting temperature agarose
156

Variations in PFGE
Orthagonal-Field Alternation Gel
Electrophoresis
Field-Inversion Gel Electrophoresis
Transverse Alternating Gel
Electrophoresis
Programmable Autonomously
Controlled Electrode
Contoured-Clamped
Homogeneous Electric Field
CHEF Electrodes

Large DNA fragments are easily broken
by shear forces. To avoid breakage of chromo-
some-sized DNA fragments, whole cells are
embedded into agarose blocks. The cells are
lysed in the agarose block and subjected to
electrophoresis. The restriction digests can also
carried out in the agarose blocks before electro-
phoresis. Several electrophoretic variables can be manipulated and these will affect the resolu-
tion. The electrophoresis conditions will depend upon the particular apparatus being used and
the size range of DNA fragments to by separated.
Factors Affecting Resolution
uniformity of the two electric fields
lengths of the electric pulses
the ratio of the lengths of the pulses
the angles of the two electric fields
strengths of the electric fields
157
CHAPTER 19--HYBRIDIZATION AND BLOTTING TECHNIQUES

DNA can be separated into single strands by
disrupting the H-bonds which hold the complementary
strands together. Complementary DNA strands can
reform the double-stranded molecule in vitro. This
ability to reform dsDNA molecules, or hybridization,
allows for the detection of specific DNA fragments or
RNA molecules. Hybridization can be carried out in
solution or after target DNA has been transferred to a
membrane support. Nucleic acids immobilized on a
membrane support can be identified with a labeled DNA
probe by blotting techniques.

In 1975 Dr. Southern described a technique to detect
specific DNA fragments after gel electrophoresis and the tech-
nique became known as Southern blotting (Box). Others modi-
fied the technique to detect specific RNA fragments and this
method became known as Northern blotting. [When protein
blotting was developed it became known as Western blotting.]
Nucleic acids can also be bound to membranes without prior electrophoresis and this process is
referred to as dot blots.

GENERAL PROCEDURES

The first steps of nucleic acid blotting are to
isolate the nucleic acid of interest and prepare it for
analysis. For example, DNA is usually digested with
restriction enzyme(s) before electrophoresis. The sample
is then subjected to gel electrophoresis on agarose gels.
In the case of RNA or ssDNA the gels are usually
carried out under denaturing conditions to minimize
secondary structures.

An optional depurination step is sometimes carried out following the electrophoresis,
especially if very large fragments of DNA are being analyzed. This step is a brief incubation
of the gel in a dilute HCl solution followed by an incubation in a neutral buffer. The brief
acid treatment will randomly remove some purine bases and lead to random breaks in the
DNA. The smaller fragments of DNA are more efficiently transferred from the gel to the
membrane.

Double-stranded DNA is denatured (i.e., the complementary strands separated) by
treating the gel with a NaOH solution. The gel is neutralized by incubating in the
appropriate buffer before transferring the denatured DNA to membranes. It is not necessary
to carry out the denaturation step if RNA or ssDNA has been electrophoresed under
denaturing conditions.
Southern
Blot

DNA on
Gel
Northern
Blot

RNA on
Gel
Dot Blot

No Gel
Generic Blotting Protocol
1. Digest DNA or isolate RNA
2. Electrophoresis
3. Depurinate (optional)
4. Denature dsDNA
5. Transfer to membrane
6. Fix nucleic acid
7. Prehybridize
8. Incubate with probe
9. Wash
10. Detect (autoradiography)
158


Transfer of nucleic acids to membrane supports. Four methods for transferring nucleic acids to
membrane supports have been described. The original method developed by Southern uses
capillary action (Figure). In this method buffer being drawn through the gel carries the nucleic
acid. The nucleic acid then binds to the membrane. The method is still widely used but takes
several hours and typically is carried out overnight. Electrophoretic transfers are also possible
but not widely used. Special devices using either pressure or vacuum have been developed. The
vacuum blotting apparatus (Figure) is the more popular and can transfer nucleic acids in
approximately 30 minutes.


Several types of membranes are used in nucleic acid blotting techniques. Nitrocellulose
has been largely replaced by nylon as the membrane of choice. Nylon is a more durable
membrane and binds more nucleic acid. Special nylon membranes, which have been modified
to contain positive charges, bind even more DNA.

Fixation of the nucleic acid to the membrane improves the sensitivity by preventing loss
of target nucleic acid during the subsequent steps. The membranes can either be 'baked' (heated
159
to 80
o
for 2 hr) or cross-linked with UV radiation using a special apparatus. Membranes are then
'prehybridized' to block non-specific binding sites. The prehybridization solution contains non-
specific DNA, such as herring sperm or calf thymus DNA. Following the prehybridization, the
membrane is incubated with the 'probe' and then washed extensively. The bound probe is then
detected by autoradiography if radioactive probes were used, or by ELISA type procedures if
non-radiocative probes (see below) were used.

It is also possible to strip the blot of the probe by incubating under conditions which do
not allow for hybridization (eg., boiling in low ionic strength). The blot can then be reanalyzed
with a different probe. This is a convenient method to compare the expression of two different
genes by Northern blotting. However, the stripping and rehybridization can only be carried out
a few times.

FACTORS AFFECTING HYBRIDIZATION

Several factors affect the binding of a DNA probe to the
target nucleic acid (Box). All of these factors affect the number or
stability of the H-bonds formed between complementary strands.
Heating DNA destabilizes the H-bonds and increasing the temperature will make it more
difficult for the probe to bind to the target DNA. The hybridization of DNA fragments is
impeded by electrostatic repulsion due to the negative charge of the phosphate backbone.
Cations neutralize this charge-charge repulsion and adding salt will promote the interaction
between probe and target. The maximum effect is achieved with 1 M Na
+
. Chaotropic agents
affect H-bond stability and therefore will reduce the hybridization of probe to target DNA. The
length and the homology of the probe to the target DNA will determine the total number of
H-bonds formed between the target and the probe. However, the total probe length tends to only
have a major affect when oligonucleotides are used as probes. DNA with a higher percentage of
GC hybridizes will form more H-bonds since GC pairs have 3 H-bonds and AT pairs have 2
H-bonds. All of these factors will play a role in the overall stability of the duplex formed
between the target and the probe.

The separation of DNA strands is sometimes called melting and exhibits a melting
temperature, or T
m
. The T
m
reflects the overall stability of any particular DNA duplex. The
effective T
m
for any particular duplex will be determined by the summation of these various
factors that affect DNA hybridization (Box). For example, the following formula can be used to
approximate the T
m
under different hybridization conditions:

Effective T
m
= 81.5
o
+ 16.6log[Na
+
] + 0.41(%GC) -
0.72(%formamide) - 1.4(%mismatch)

Consistent with the above discussion, increasing the sodium concentration and percentage of
GC base pairs will raise the T
m
and increasing the formamide (a chaotropic agent) concentration
and decreasing the amount of homology between the probe and target will lower the T
m
. Often
the %mismatch is not known and it not used in the calculation. In the case of synthetic
oligonucleotides the absolute probe length and amount of GC base pairs are the major factors
involved in hybridization. Thus, the following formula results in a more accurate estimation of
temperature
ionic strength
chaotropic agents
probe length
probe mismatch
% GC
160
T
m
:

Effective T
m
= 2(A + T) + 4(G + C)

where A, T, G and C refer to the absolute numbers of each of the nucleotides.

Generally, hybridization is discussed in terms of strin-
gency and not the T
m
. Stringency refers to the conditions of the
hybridization. It is a relative term that is related to the T
m
(Box)
and reflects the homology between the probe and the target. For
example, only a probe with a high degree of homology to the
target DNA will hybridize under high stringency conditions, whereas low stingency will allow a
less homologous probe to hybridize to the target DNA.

The stringency is controlled by changing the hybridization conditions. For example,
increasing the temperature, decreasing the salt concentration, or including formamide all
increase the stringency. Likewise, the stringency can be decreased by lowering the temperature,
increasing the salt concentration, or decreasing the formamide concentration. In practical terms
it is often easier to vary the sodium or formamide concentrations rather than the temperature. A
common buffer for hybridization is SSC (standard sodium citrate) which is a citrate buffered
sodium solution. A stock solution of 20X SSC (= 3.3 M Na
+
) is diluted to achieve different
levels of stringency. Typically ranges of 0.1-6X SSC are used with the lower SSC
concentrations representing higher stringency. The formamide is typically used to lower the
temperature and at the same time maintain a certain level of stringency. For example, hybridiza-
tions carried out at 60-65
o
in the absence of formamide or at 37-42
o
with 50% formamide are
about equal in stringency. If incubations at high temperatures are inconvenient then formamide
can be included in the buffers and the hybridizations carried out at lower temperatures.

Stringency also applies to the wash steps. It is common to hybridize at low stringency
and then to wash at a higher stringency. This insures that the probe will bind to the target DNA
and any non-specific hybridization can be removed during the wash steps. The conditions for
hybridization (i.e., stringency) need to be determined empirically in conjunction with the above
formula as guidelines. A single blot can be sequentially examined under different stringencies.
Hybridization and washes are initially carried out at low stringency. The blot is wrapped in
plastic and not allowed to completely dry. After exposure to x-ray film, the blot is washed under
higher stringency and reexposed to X-ray film. Comparison of the different autoradiographs
will allow one to determine how homologous the probe is to the target DNA and the degree of
cross-hybridization to other DNA fragments.

PREPARATION OF LABELED DNA PROBES

The two major sources of probes are previously cloned
genes and synthetic oligonucleotides. In both cases a label needs
to be incorporated into the probe DNA. Radioactivity is a common label, but non-radioactive
probes are also available. Four methods for incorporating label into DNA probes have been
described (Box ). Nick translation is an older technique that has been replaced by random
Stringency vs. T
m

high T
m
- 15
o
moderate T
m
- 25
o
low T
m
- 35
o
Nick Translation
Random Priming
T4 Nucleotide Kinase
Terminal Transferase
161
priming. Random priming is the method of choice for labeling cloned DNA fragments.
Synthetic oligonucleotides are labeled using T4 nucleotide kinase.

Random Priming.

In random priming (Figure) DNA is
denatured by heating and mixed with hexamers
of random sequence (i.e., random primers). The
random primers are usually synthesized and
included as part of a kit. They can also be
prepared from genomic DNA. A few of primers
will be complementary to the probe DNA and
the duplex formed between the primer and the
probe DNA will serve as an initiation point for
the DNA polymerase. The DNA polymerase
used is Klenow. Klenow is the large subunit of
DNA polymerase I in which the 5'3'
exonuclease activity is removed. The four
dNTPs including a nucleotide containing a radioactive phosphate in the -position are also
added to the mixture. Therefore, the Klenow will make
radioactive copies of the template DNA. The probe
DNA is boiled immediately before use in the
hybridization assay to convert the dsDNA to ssDNA.

T4 Polynucleotide Kinase.

T4 polynucleotide kinase transfers the -PO
4

from ATP to the 5'-hydroxyl of polynucleotides. It is
therefore necessary to dephosphorylate the DNA with
alkaline phosphatase (AP) before carrying out the
phosphorylation. A disadvantage of this technique is
that only one radioactive atom is incorporated per
DNA strand. However, 5'-terminal phosphorylation is widely used to label oligonucleotide
probes that have been prepared synthetically. Synthetic oligonucleotides lack the 5-phosphate
and are too short for random priming. T4 kinase is also used in Maxim and Gilbert DNA
sequencing and to phosphorylate (non-radioactive) synthetic linkers.

Terminal Transferase.

Terminal deoxynucleotide transferase (TdT) adds
dNTPs to the 3-OH of either ssDNA or to 3' overhang. In
the presence of Co
2+
TdT will add dNTPs to the 3'-OH of
either dsDNA or 5' overhangs. TdT can be used to
radiolabel 3' ends if radioactive nucleotides are used. A
more common use, however, is to generate homopolymer tails for molecular cloning.
162
Non-radioactive Probes.

Several procedures have been devised for the
detection of hybridization using non-radioactive
probes (Box). All are based upon enzyme-linked
systems using either alkaline phosphatase (AP) or
horse-radish peroxidase (HRP). Biotinylated dNTPs
can be incorporated into the probe DNA by random
priming. The probe is then be detected with an
enzyme-linked streptavidin. Another approach is to
incorporate digoxigenin-11-(d)UTP into the DNA
probe and then subsequently detected with enzyme-
linked antibody against the digoxigenin. A
third method is to directly cross-link HRP
to the DNA probe.

Radioactive probes are generally
more sensitive and reliable. However,
non-radioactive probes can be adapted to
many applications and eliminate some of
the problems associated with the use of radioactivity such as waste disposal and safety issues. In
addition, the use of non-radioactive probes is are particularly advantages in situation where the
same probe is going to be used over a long period of time. The short half-life of
32
P (14 days)
necessitates that the probe be prepared on a monthly basis, whereas large amounts of a non-
radioactive probe can be prepared and stored for long periods of time.

Insoluble substrates, as described for Western blots, or chemiluminescent substrates can
be used in association with Northern and Southern blots. Chemiluminescent substrates produce
light when cleaved by the appropriate enzyme and this light is detected by autoradiography.
Substrates for both alkaline phosphatase (1,2 dioxetane) and peroxidase (luminol) are available
(figures). The use of chemiluminescence allows the blot to be striped and reprobed.




Labeling/Detection Systems
Biotin/Streptavidin
Digoxigenin-UTP/Antibody
Enzyme Crosslinking

163
RFLP AND RESTRICTION MAPPING

Restriction fragment length polymorphisms
(RFLP) refers to the gain of loss of restriction sites
associated with a genetic locus. These polymorph-
isms can be used to distinguish species, strains or
even individuals. Single nucleotide mutations can
result in the loss or gain of restriction sites, and thus
generate different sized DNA fragments associated
with that locus. The typical procedure is to digest the
target DNA with a restriction enzyme(s), separate the
fragments by gel electrophoresis, and to detect the
fragments of interest by Southern blotting (Figure).
The changes causing the size polymorphisms do not
have to be directly associated with the genetic locus
being used for a probe.

RFLP applications range from determining
specific mutations associated with a particular gene
to a genetic 'fingerprinting'. The optimal restriction
enzyme(s) and probes will need to be empirically determined for a particular application. A
limitation for the application of RFLP is that relatively large amounts of highly purified DNA
are needed for many applications. In addition, the procedure is laborious and can take several
days to carry out.

Restriction Mapping. Genetic loci can be defined by restriction maps which show the
relative positions of restriction sites. The maps can be generated from sequence data or
determined experimentally when sequence data is not available. Restriction maps are generated
by digesting DNA with different combinations of enzymes and analyzing the digests by gel
electrophoresis or Southern blotting. The sizes of the fragments will indicate the distant between
restriction sites. The general procedure is to digest the sample DNA with individual restriction
enzymes and combinations of the enzymes. The change in the size of the restriction fragments
in double digests is used to determine the relative positions of different restriction sites.

For example, in the situation on the left (see figure below) digestion with enzyme A
produces and single 10 kb fragment and digestion with enzyme B produces a single 8 kb
fragment. Carrying out a double digest with a mixture of enzymes A and B would result in the
same 8 kb fragment as digesting with enzyme B alone. This indicates that both of the B
restriction sites are between the A restriction sites, but no information about the distance
between site A and site B is obtained. If the situation is as depicted on the right, then a fragment
that is < 8 kb is produced. The size of this fragment will indicate the distance between the A and
B restriction sites. A more precise map can be generated in both situations if all of the
restriction fragments are detectable.

164


IN SITU HYBRIDIZATION

Cell or tissue specific gene expression
can be determined by in situ hybridization.
Tissues are fixed on slide and hybridized with
cDNA probes. The cells expressing the gene of
interest are detected by autoradiography. The
cells expressing the gene of interest are identi-
fied by the reduced silver grains (dark areas in
figure) after development of the photographic
emulsion. The probes are usually made with
35
S rather than
32
P. The probes can also be
labeled with fluorochromes and detected by
fluorescent microscopy. In situ hybridization is
commonly used in developmental biology to
determine when and where developmentally
regulated genes are expressed.
165

DNA MICROARRAYS

Most of the hybridization methods are designed to analyze one gene at a time. The DNA
microarray technology, also called gene chips, genome chips, biochips, DNA chips, provides a
means to simultaneously analyzed thousands of the genes. The array is an orderly arrangement
of known DNA samples. Sample spot sizes in macroarrays are generally > 300 microns in
diameter and can be imaged with conventional gel and blot scanners. Microarrays contain
thousands of spots that are generally less than 200 microns in diameter. Preparation of
microarrays requires specialized robotic equipment and analysis of the microarrays requires
special imaging equipment.

The DNA microarray is fabricated by spotting known DNA samples, or 'probes', on a
solid support such as glass or nylon through the use of high speed robotics. The probes can
either be known cDNA sequences (500-5000 bases) or synthetic oligonucleotides (20-80
nucleotides). The array is then exposed to flourescent-labeled cDNA prepared from total
mRNA. This free 'target' DNA will bind to the spots containing homologous sequences. The
identity and abundance of the complementary sequences can then be determined. It is also
possible to mix DNAs prepared with different fluorescent labels. Signals are measured as
absolute intensities for a given target, or as ratios of two probes with different fluorescent labels
representing two separate treatments to be compared or with one probe as an internal control.

http://www.bsi.vt.edu/ralscher/gridit/intro_ma.htm


167
CHAPTER 20--POLYMERASE CHAIN REACTION

One problem with detecting specific DNA sequences,
especially those of unique genes, is that a relatively large amounts
of pure DNA are needed. In addition, blotting techniques are
laborious and time consuming to carry out. To circumvent these
problems it is possible to enzymatically amplify a specific region
of DNA using the polymerase chain reaction (PCR). The ability
to detect minute amounts of specific DNA sequences has resulted
in the broad application of PCR for diagnosis, forensics,
molecular epidemiology, etc. In addition, PCR is an integral
aspect of many methods, such as gene cloning and sequencing.

PCR MECHANISM

PCR amplifies a specific segment of
DNA that lies between two known primer
sequences. DNA strands are separated by
heating and then annealed with a pair of
primers which are complementary to the
opposing strands (Figure). DNA polymerase
recognizes this small region of duplex DNA as
a substrate and in the presence of nucleotides
will synthesize the complementary strands of
both template strands. If this procedure is
repeated, the newly synthesized fragments also
serve as templates for subsequent rounds
resulting a geometric amplification. The end
product of PCR is a dsDNA molecule that is
defined by the 5'-ends of the primers. In other
words, the length of the DNA molecule is
determined by the distance between the
primers.

Typically, PCR results in a million-fold
amplification of the target DNA. Therefore,
sequences that only represent a small
proportion of the total DNA can be detect after
PCR and generally the DNA does not have to be highly purified. Furthermore, the procedure is
relatively easy to carry out and does not require expensive equipment or reagents.

One of the major factors in the success of PCR is the use of DNA polymerases isolated
from thermophilic organisms such as Thermus aquaticus. Taq polymerase is thermostable and
exhibits an optimal temperature for activity at 72-74
o
. Other thermostable DNA polymerases are
also available. The temperature necessary for the separation of DNA strands will destroy the
DNA polymerase activity from other sources. In addition, the optimal temperature for activity
PCR Applications
Diagnosis
Taxonomy
Forensics
Molecular
Epidemiology
Gene Expression
Sequencing
Gene Cloning
Probe Generation
Site-Directed
Mutagenesis
168
(i.e., 37
o
) would result in low stringency hybridization leading to the synthesis of irrelevant
fragments. The need to change the temperature during the PCR procedure (i.e., denaturation,
primer annealing, polymerization) requires a thermocycler. The thermocycler is an instrument
that rapidly and accurately changes the temperature of a metal block designed to hold 0.5 ml
microcentrifuge tubes.

PRACTICAL ASPECTS

The first step of PCR is to combine the template
DNA, primers, dNTPs, Mg
2+
and Taq polymerase in a
single tube (Box). Special thin-walled tubes of uniform
thickness are used to ensure rapid and equal temperature
changes throughout the reaction volume. Typical concentrations are approximately 0.1 pM of
the target DNA, 2 nM of primers with a T
m
> 55
o
, 20-200 M dNTPs (lower concentrations
lessen mispriming), and 2 units of enzyme. The optimal Mg
2+
concentration depends upon the
total dNTP concentration which includes: free dNTPs, primers, and template DNA. Since the
template DNA is the most variable, it is generally necessary to titrate the optimal Mg
2+

concentration for different sources of templates. In addition, the EDTA (a typical component of
buffers used for nucleic acid isolations) will affect the free Mg
2+
concentration.

The thermocycler is programmed for the desired times and temperatures of denature-
tion, annealing and polymerization (Table). The optimal conditions are determined empirically.
Most thermocyclers can also be programmed to vary these parameters in different cycles. For
example, the denaturation step in the first cycle is sometimes carried out for 2-5 minutes to
ensure a more complete melting of the target DNA. Likewise, after the last polymerization
cycle it is possible to lower the temperature to 4
o
(i.e., storage conditions).

PCR Parameters
STEP TEMP TIME NOTES
Denature 94-96
o
0.5-2 min
longer times denaturation, but
enzyme and template
Annealing 15-25
o
< T
m
0.5-2 min
higher temp. and shorter times
specificity, but yield
Polymerization 72-75
o
~1 min (<kb)
Taq processivity = 150
nucleotides/sec

The oligonucleotide primers are the most
critical element in terms of successful PCR (Box).
Computer programs are used to examine DNA sequen-
ces for potential primers. One consideration is distance
between the primers. Smaller DNA fragments are
amplified more efficiently than longer DNA fragments
and it is often particularly difficult to amplify frag-
ments larger than one kb. The primers should also be
unique in that they should not hybridize to other sites
in the target DNA. Generally primers of 20 nucleotides or greater will provide a satisfactory
PCR Procedure
1. Mix DNA, primers, dNTPs
and Taq polymerase.
2. Set the thermocycler for
desired times.
3. Analyze amplified DNA.
Primer Design
template length
uniqueness (18-28 bases)
no primer dimers
no internal complementarity
50% GC ratio
3'-GC 'caps'
HPLC purification
169
level of uniqueness. The primers should not exhibit complementarity to each other or internal
complementarity within a primer. Both of these will phenomenon will prevent the primer from
annealing to the template DNA. A balance nucleotide composition also tends to improve the
function of a primer. Designing a primer so that the last one or two bases on the 3'-end are either
a G or a C will result in a stronger hybridization and ensure recognition by the polymerase.
Purifying the primers by HPLC will improve their fidelity, but is not absolutely necessary.

Amplified DNA is analyzed by gel electrophoresis. In most applications the desired
result is a single band, or amplicon, detectable with ethidium bromide. Primers are usually
designed so that a sample can be tested for the presence or absence of the expected band. Size
heterogeneity can be determined if exhibited by that particular locus. Southern blotting of the
PCR products can also be carried out for either increased sensitivity or specificity.


Generally 25-30 cycles will yield the maximum amount of amplification. This is due to
loss of enzyme activity associated with the high temperatures and to the accumulation of pro-
ducts that inhibit or interfere with the amplification. Such products include fragments synthe-
sized as a result of mispriming and short fragments formed as a result of the polymerase
dissociating from the template before reaching the opposite primer. Subjecting the sample to a
second round of amplification can increase the sensitivity. Typically the reaction product from
the first cycle is diluted by a 100-1000 fold and the primers, nucleotides and polymerase are re-
added. Similarly, a nested PCR can be designed in which a second amplification is carried out
using a primer pair internal to the first set of primers (Figure).




Primers for Nested PCR
Site Directed Mutagenesis
PCR can also be used to manipulate DNA. For example, site-directed mutagenesis
can be carried out by designing primers with single nucleotide mismatches. Since
the primers serve as templates in subsequent rounds of DNA replication the PCR
products will contained the introduced nucleotide. Similarly, restriction sites are
easily added to the PCR products for subsequent subcloning as illustrated by the
following figure:

170
RNA-PCR

It is also possible to use PCR for the analysis of RNA and gene expression. This
method, called RNA-PCR or RT-PCR, detects specific transcripts and exhibits a greater
sensitivity than Northern blotting. The first step of RT-PCR is to make a DNA copy of the
mRNA (i.e., cDNA). The copy is made using reverse transcriptase (RT), an enzyme of
retroviruses which exhibits RNA-dependent DNA polymerase activity. There are three basic
strategies for the synthesis of cDNA: specific priming, oligo-dT priming, or random priming
(Figure). The cDNA is then subjected to PCR using a specific primer pair.




QUANTITATIVE PCR

Information obtained from PCR is primarily qualitative. The major problem is that
the analysis is an end-point determination. Since the inhibitory affects accumulate the
differences in the initial template concentration will be masked. In addition, quantification
of the end-point amplification product usually requires some post-amplification handling
which increases the labor and time of the assay.

One approach to obtaining quantitative information is to include a competitive
reporter with the template DNA. This reporter has the same primers as the target DNA, but
produces a different size amplicon. The target DNA is mixed with increasing known
amounts of the competitor and analyzed by PCR and gel electrophoresis. Since the reporter
and the target DNA are in the same tube, they will be similarly inhibited. Therefore, the
amount of the respective amplicons produced will be proportional to the initial template
concentrations. Approximately equal amounts of competitor and target amplicons will be
produced when their respective template concentrations are equal, and therefore, the amount
of target DNA (i.e., number of copies) can be estimated from the amount of reporter known
to be in that sample.

Another method for estimating the concentration of target DNA is by 'real time'
171
PCR. Real time PCR utilizes fluorescent probes for the detection of PCR products and
requires a special thermocycle (i.e., 'light cycler') which monitors fluorescence during the
reaction. The simplest method to monitor the accumulation of PCR products is to include a
dye in the PCR reaction which fluoresces when intercalated into dsDNA. SYBR green is the
most widely used dye for this application. The dye binds to the double-stranded PCR
products and fluoresces. Fluorescence is measured during the extension stage of each cycle
and thus represents the the accumulation of PCR products during the reaction. The cycle in
which significant fluorescence is first detectable is inversely proportional to the amount of
initial template. A standard curve can be prepared from this threshold cycle (C
t
) and the
template concentration of known standards. The amount of template present in an unknown
sample can then be extrapolated from this standard curve.


Since these dyes bind to any dsDNA, specificity can be a problem. This is especially
true towards the end of the PCR cycles when more non-specific products may be
accumulating. Furthermore, it is not possible to quantify multiple PCR products within the
same sample (i.e., multiplexing). Several single-stranded DNA detection methodologies
have been developed to overcome these limitations. Two examples include hydrolysis
probes (aka, F-Q probes or Taqman probes) and molecular beacons. Both of these methods
rely upon the use of specific probes to the amplified sequence which have been coupled with
fluorochromes and quenchers, thus adding some expense to the these assays. Through the
use of fluorochromes with different emission spectra it is possible to develop multiplex
assays in which differenent PCR products can be monitored simultaneously in the same
sample.

Hydrolysis probes. A probe homo-
logous to a sequence between the two
primers is designed. One end of the probe
contains a fluorochrome (eg., 5') and the
other end contains a quencher (eg., 3').
The energy of the excited fluorochrome is
transferred directly to the quencher and
therefore this probe does not fluoresce.
During annealing the probe also
hybridizes to the target DNA. As the
polymerase progresses down the template during the extension reaction the endogenous 5'-
172
exonuclease activity digests the F-Q probe and thereby releases the fluorochrome and
quencher. The fluorochrome and quencher are no longer held together in close proximity
and therefore the free fluorochrome fluoresces. Free fluorochrome accumulates after each
cycle and the fluorescence intenesity is a measure of the amount of specific PCR product.
There tends to be a high background associated with the F-Q probes since the fluorochrome
and quencher are separated by approximately 20 nucleotides.

Molecular beacons. Probes which form 'hiarpins'
have been designed to improve the intramolecular
quenching. As in the case of the F-Q probes a probe
homologous to a sequence between the primers is
designed. Complementary sequences are added to each
end as well as the the fluorochrome and quencher. These
complementary sequences will anneal and therefore
force the fluorochrome and quencher in close proximity
and thus increase the efficiency of quenching.
Hybridization of the probe to the target DNA results in
fluorescence since the fluorochrome and quencher are
now held far apart. As in the case of using non-specific
dsDNA binding dyes, the fluorescence is only observed
during the annealing step. During the extension reaction the probe is displaced and will be
recycled.

Special considerations are needed for probe and primer design in both the Taqman
and molecular beacon assays. The optimal amplicon length for both assays is 50-100 bp. In
both cases the probe needs to bind to the target in a quantitative manner and with the desired
specificity. The T
m
of the probe-target hybrid should by 8-10
o
C above the annealing temper-
ature of the PCR reaction. This is usually obtained with probes of 20-26 nucleotides. The
exact length will depend on GC content and the amount of tolerance for mismatches. The
stem region for the molecular beacon probes should also exhibit a T
m
of 8-10
o
C above the
annealing temperature so that unhybridized probe is kept in the non-fluorescent conforma-
tion during the annealing step. Stems of six nucleotides, of which five are G:C pairs are
usually the optimum.

LIMITATIONS

One limitation of PCR is that only DNA between two known
primer sequences is amplified, thus making it difficult to analyze
unknown sequences. Modifications of the PCR method which overcome this limitation have
been described (Box).

Inverse PCR (also known as chromosome crawling) is a method to amplify the
sequences flanking a known sequence. Genomic DNA is digested with a restriction enzyme
which produces appropriately sized fragments (eg., 3-5 kb) containing the target sequence. The
fragments are circularized with DNA ligase and amplified with primers that are directed away
from each other. This results in the amplification of the sequences that directly flank the known
inverse PCR
anchor PCR
RAPD
AFLP
173
sequence. The boundary between the two flanking sequences is demarcated by the restriction
site used to disgest the genomic DNA.

Known sequences, or 'anchors', can be also added to
unknown sequences and used as the source of PCR primers.
This anchor PCR is often used to amplify the unknown 5'-end
of a cDNA (i.e., mRNA). (Sometimes called RACE for rapid
amplification of cDNA ends.) In this method a cDNA copy of
the mRNA is made using a primer based on known sequence
(P1). A homopolymer tail is added to the 3'-end of the cDNA
with terminal deoxynucleotide transferase. A primer
containing sequence complementary to the homopolymer tail
(P3) is used in conjuction with a primer from the known
sequence (either P1 or P2) to amplify the target DNA.

Random portions of the genome can also be amplified
as a means to identify polymorphisms without prior
knowledge of the particular polymorphism. For example,
random amplified polymorphic DNA (RAPD), also called
arbitrarily primer PCR (AP-PCR), uses short (usually 10
nucleotides) random sequence primers to amplify random
portions of the genome. These short primers will
anneal to a number of locations within a genome
and in loci with inverted repeats the appropriate
distance apart an amplicon will be produced. The
fingerprint of DNA fragments can then be
compared between species or strains and
analyzed for polymorphisms. The RAPD method
has several advantages as an assay for
polymorphism. However, it does tend to exhibit
poor reproducibility.

Another method based on the amplifica-
tion of random portions of the genome is ampli-
fied fragment length polymorphisms (AFLP). The
advantage of this method over RAPD analysis is
that the polymorphisms are based on restriction
sites and not primer binding. Therefore, AFLP
analysis tends to be more reproducible than RAPD
analysis. The first step in AFLP analysis is to
digest genomic DNA with two restriction
enzymes. Typically one of the enzymes has a 6-
base recognition sequence and the other a 4-base
recognition sequence. The DNA is then ligated with adaptors corresponding to the two
restriction sites and PCR using primers based on the adaptor sequences are carried out in the
Advantages of RAPD
requires small amounts of DNA
(15-25 ng)
no prior sequence information
needed
easily produced and screened
relatively easy to convert into
more reproducible PCR assay
174
presence of either radioactive nucleotides or fluorochrome-labeled nucleotides. The PCR
products are analyzed on sequencing gels for polymorphic fragments. Both AFLP and RAPD
are used to identify strain or species specific markers similar to RFLP analysis. However, much
less DNA is need and these methods are less laborious and time consuming than traditional
RFLP analysis.

PRECAUTIONS

The ability of PCR to amplify minute amounts of
DNA is not only a strength, but also a potential weakness.
Under controlled situations amplification of a single copy of
DNA has been demonstrated. This ability poses a serious
problem in terms of contamination of buffers and samples
with unwanted DNA. Precautions against spurious contami-
nation need to be taken (Box). It is also advisable to include a control sample (no target DNA
added) in every PCR run.
Precautions
avoid contamination
aliquot reagents
add target DNA last
no target DNA control
prepare + control
elsewhere
175
CHAPTER 21--RECOMBINANT DNA

Molecular cloning, or recombinant DNA, is the process by which a single gene, or
segment of DNA, is isolated and amplified. This is accomplished by recombining DNA
fragments in vitro and then propagating the DNA fragments of interest. A source of nucleic
acids and the ability to manipulate them are needed to carry out molecular cloning (Box).

The source and type of nucleic acid will depend on
the gene being cloned and the species of interest. The vector,
when introduced into the appropriate host, is a replicating
piece of DNA. The foreign DNA is recombined with a
vector, which is then replicated by the host. Available
cloning vectors include plasmids, phage/viruses, or some
combination of the two. Phage, short for bacteriophage, are
viruses of bacteria.

In order to carry out recombination between the
vector and the foreign DNA, it is necessary to specifically and reproducibly cleave DNA.
Restriction enzymes cleave DNA in a site specific manner and thus provide this requirement.
DNA ligase is an enzyme that covalently joins two pieces of DNA. Therefore, restriction
enzymes and DNA ligase allow for the recombination of foreign DNA with vector DNA.

The DNA of interest is amplified by introducing the recombinant DNA (rDNA) into a
compatible host. The vector is replicated by the host. The introduction of vector in the host
depends upon the nature of the vector and host. For example, some viral vectors can be
introduced by infection. Introduction of naked viral DNA or plasmid DNA is referred to as
transfection and transformation, respectively. Finally a means to detect the molecular clone
of interest, or a screening mechanism, is needed.

PREPARATION OF FOREIGN DNA

The two major sources of foreign DNA for molecular cloning are genomic DNA
(gDNA) and complementary (or copy) DNA (cDNA). Complementary DNA is prepared by
making a copy of mRNA. One consideration in deciding whether to use gDNA or cDNA is the
copy number and the level of expression of the gene of interest. Single copy genes in large
genomes may require extensive screening. A gene expressed at high levels, though, will require
less screening to isolate a cDNA clone. Conversely genes expressed at low levels will be
difficult to detect in cDNA libraries. In addition, some genes may only be expressed at certain
times or in certain tissues that are inaccessible or unavailable, thus necessitating gDNA
libraries.

One also needs to consider how the recombinant DNA library will be screened and the
questions that will be asked about the cloned gene. For example, many eukaryotic genes contain
'introns' which will result in an interruption of the open reading frame. Therefore, cDNA is
usually a better starting material if an expression library is going to be screened or if the cloned
gene is going to be used for the production of recombinant protein. Similarly, comparing gDNA
Cloning Needs
foreign DNA
vector + host
means to cleave DNA
ability to join DNA
fragments
introduction of rDNA
into host
screening mechanism
176
and cDNA will allow for a precise mapping of the introns and splice sites.

Fragmenting gDNA. The extreme length of gDNA
necessitates that it be fragmented into smaller pieces before
molecular cloning (Box). The size of the DNA fragments
should take into account the vector capacity and the size of the
desired clone. Restriction enzymes reproducibly cleave DNA
at specific sites which are relatively easy to ligate into
restriction sites of vectors. If some randomness is needed (eg., expression libraries and
overlapping clones), it may be possible to carry out the digestions under 'star' conditions or to
use frequently cutting enzymes under conditions where the target DNA is only partially
digested.

Preliminary Southern blots are often carried out when a DNA probe is available. The
gDNA is digested with several different enzymes and analyzed for appropriate sized fragments.
The appropriateness of the size depends on the size of the genetic locus and the vector being
utilized. After deciding which restriction enzyme(s) is the most appropriate, a sufficient quantity
of DNA is digested and electrophoresed. The region of the gel corresponding to the desired size
range is excised and the DNA isolated. This size-fractionated DNA is then used to prepare a
library.

It is also possible to use a non-specific nuclease under controlled conditions to cleave
gDNA into appropriately sized pieces. For example, DNAse I in the presence of Mn
2+
will
cleave dsDNA at approximately the same site on both strands. Pilot experiments are carried out
to determine the optimal DNase/DNA ratios for the production of the desired sized fragments.
An application of this method is in 'shotgun' sequencing strategies which result in the
production of several overlapping clones. Mung bean nuclease in the presence of formamide
will also randomly cleave dsDNA.

DNA can also be broken by mechanical shear forces. Passage of a DNA solution
through a small gauge syringe needle or sonication will result in the DNA molecule being
broken in random positions. Following fragmentation by either method the DNA can be size
fractionation by either rate zonal centrifugation on sucrose gradients or by gel electrophoresis.
Restriction sites can be added to the ends of the gDNA fragments to increase the efficiency of
cloning (see below) as compared to blunt-end ligations.

PCR fragments can also be cloned. Amplifying DNA fragments of interest before
cloning will eliminate, or at least reduce, the screening of recombinant DNA libraries. However,
the cloning of PCR fragments is less efficient than anticipated due to the non-template
dependent addition of a single nucleotide to the 3'-end of PCR products by many thermostable
DNA polymerases. The most frequently added nucleotide is an adenosine residue. Vectors
containing a single 3'-T overhang (i.e., TA cloning) greatly enhances the efficiency of cloning.
In addition, it is possible to generate a restriction site during the amplification process (see
PCR).

Fragmenting gDNA
restriction enzymes
random nucleases
mechanical shearing
177
Preparing cDNA Libraries. The preparation of cDNA
from mRNA is a multistep process (Box). The first step in
preparing cDNA is to isolate mRNA. mRNA tends to be less
stable than DNA and requires more precautions than gDNA.
In addition, an enrich source of mRNA of interest can be
used. Enrichment for a particular mRNA could include:
choice of species, tissue, developmental stage, induction of
expression (eg., hormones), etc. It may also be possible to
enrich for the mRNA of interest by size fractionation,
immunological purification of polysomes, or substractive libraries. The integrity of mRNA
should be evaluated by Northern blotting and/or in vitro translation before proceeding with the
preparation of the library.

After obtaining the desired mRNA a DNA-RNA hybrid is synthesized with reverse
transcriptase (RT). Reverse transcriptase is a RNA-dependent DNA polymerase isolated from
retroviruses. In the presence of a mRNA template and primer, RT will copy the RNA template
resulting in the formation of a DNA-RNA hybrid. An oligo-dT primer of 12-18 bases in length
is used to initiate synthesis on poly-A messenger RNA. Sequences corresponding to restriction
sites can also be included in the oligo-dT primer. For example, a common primer-linker for the
synthesis of cDNA is 5'-(GA
10
)ACTAGTCTCGAG(T)
18
-3' which includes SpeI and XhoI sites.
Random priming can also be used to increase the chances of obtaining the 5'-end of long mRNA
or if the mRNA of interest lacks a poly-A tail.


Three Methods for the Synthesis of cDNA 2nd Strands

The RNA strand is replaced with a second DNA strand following the synthesis of the
RNA-DNA hybrid. Three methods for synthesizing the 2
nd
DNA are: 1) self-priming, 2)
Preparation of cDNA
1. Isolate mRNA
2. Synthesize DNA-RNA
hybrid
3. Synthesize 2
nd
DNA
strand
4. Add termini
178
replacement synthesis, and 3) primed synthesis (Figure). In self-priming the RNA-DNA hybrid
is denatured by heating or the RNA is hydrolyzed with NaOH. The 3'-termini of single-stranded
cDNA forms a hairpin loop that can function as a primer for DNA polymerase. Following the
completion of second strand synthesis, the ssDNA loop is then removed with S1 nuclease. The
self-priming is a poorly controlled reaction and treatment with S1 nuclease sometimes leads to
loss of the 5'-end of the message.

Replacement synthesis uses the DNA-RNA hybrid as a template for a 'nick
translation' reaction. The DNA-RNA hybrid is treat with RNAse H which produces nicks in
the RNA strand. The nicked RNA then functions as primers for DNA polymerase. In primed
synthesis, a homopolymeric tail (or other known sequence is added to the 3'-end using
terminal transferase (TdT). A primer complementary to the homopolymer tail is then used to
synthesize the second strand. This method may result in a higher efficiency of recombinant
clones containing the complete 5'-end of the message.

The synthesis of cDNA results in blunt-end DNA
without convenient restriction sites for cloning into
vectors. Linkers are added to the termini of the cDNA
molecules which will provided the necessary restriction
sites (Figure). Restriction sites within the insert DNA are
protected by methylation prior to the addition of the
linkers. The overhangs are then generated by digestion
with the appropriate restriction enzyme. The use of primer-
linkers allow for different restriction sites on the 5'-end and
the 3'-end (i.e., poly-A tail) and directional cloning.
Another strategy to create sticky ends is to add
complementary homopolymer tails with TdT to both the
insert DNA and the vector.

PLASMID VECTORS

Recombinant DNA vectors function as carriers
of the foreign DNA. Plasmids are extra-chromosomal
genetic elements found in a variety of species and are
widely used in molecular cloning. They are autono-
mously replicating circular DNA molecules that range in size from 1-200 kb. Their replication
can be under either stringent control (low copy number) or relaxed (high copy number).
Stringent control means that the plasmid is only replicated when the genome is replicated,
whereas relaxed plasmids are replicated continuously. Many plasmids confer antibiotic
resistance and are naturally transmitted between bacteria by conjugation.

Naturally occurring plasmids have been modified to make them more useful as cloning
vectors (Box). To be useful as cloning vectors, plasmids need to be relatively small, replicate in
relaxed fashion, have selectable markers (eg., antibiotic resistance) and contain unique
restriction sites. The most widely used plasmids are derivatives of either pBR322 or pUC
(where the lower case 'p' means plasmid). Typically modifications of plasmids include the
Useful Plasmid Features
relaxed replication
selectable markers
streamlined
unique restriction sites
recombinant identification
179
removal of unnecessary portions, or streamlining, and the introduction of several useful
restriction sites in one location. These restriction sites, call the multiple cloning site (MCS) or
polylinker, facilitate DNA cloning. This MCS is usually located within a gene which allows for
easy identification of recombinants.

Recombining target DNA with vector.

The first step in cloning involves
preparing the foreign or target DNA and
simultaneously preparing the plasmid
vector (Box). The exact method for
preparing the foreign DNA will depend on
the source of the nucleic acid and the nature of the gene
being cloned. For example, genomic DNA can prepared by
digesting with appropriate restriction enzyme(s). The
digested DNA can also be size-fractionated on agarose gels
and the desired size range of DNA fragments excised from
the gel.

The vector is prepared by digesting the plasmid
with the restriction enzyme(s) that are compatible with the
foreign DNA. Recombinant DNA molecules are generated
by simply mixing the prepared vector DNA with the
prepared foreign DNA in the presence of DNA ligase
(Figure). The complementary overhangs, or 'sticky-ends',
produced by the restriction enzyme(s) allow the foreign
DNA and vector DNA to transiently anneal. DNA ligase
catalyzes the formation of a phosphodiester bond between
a 5'-phosphate and a 3'-hydroxyl (Figure) and will result in
the covalent joining of DNA fragments. Typically, equal
molar amounts of vector and foreign DNA are mixed
during the ligation reaction to minimize the formation of
recombinants with tandem copies of the insert. The molar
ratio of vector to foreign DNA can be increased (eg., 1.5-2) to increase the amount of target
DNA incorporated into plasmids. Ligation reactions are usually carried out at lower
temperatures (8-15
o
) for extended time periods (2-24 hours) to promote the annealing of the
short (2-4 base) overhangs.

Generic Recombinant DNA Protocol
1. Prepare foreign DNA.
2. Prepare vector.
3. Combine foreign DNA with vector.
4. Introduce recombined vector into host.
5. Screen for rDNA of interest.
Example of a multiple cloning site

| SacI | | ScI I | | XbaI | | SpeI | | BamH| | SmaI | | PstI | | EcRI | | EcRV| | HI I I | | ClaI | | SalI | | XhoI | | KpnI |
GAGCTCCACCGCGGTGGCGGCCGCTCTAGAACTAGTGGATCCCCCGGGCTGCAGGAATTCGATATCAAGCTTATCGATACCGTCGACCTCGAGGGGGGGCCCGGTACC
CTCGAGGTGGCGCCACCGCCGGCGAGATCTTGATCACCTAGGGGGCCCGACGTCCTTAAGCTATAGTTCGAATAGCTATGGCAGCTGGAGCTCCCCCCCGGGCCATGG
180


During DNA ligations the vector can
ligate with itself in cases were the vector has
identical overhangs on both ends (including
blunt-ends). This intramolecular formation of
circles is kinetically favored and many of the
vectors will not contain insert DNA (i.e., be
non-recombinants). The formation of non-
recombinants is minimized by treating the
vector with phosphatase prior to the ligation
step. This pretreatment favors the formation of
recombinants since DNA ligase requires a 5'-
phosphate, which can now only be supplied by the foreign DNA insert (Figure).

Commonly used phosphatases are calf intestinal alkaline phosphatase (CIAP) and
shrimp alkaline phosphatase (SAP). SAP is inactivated by moderately high temperatures
whereas CIAP is more stable and requires higher temperatures or proteinase K treatment for
complete inactivation. The phosphatase activity needs to be completely inactivate before mixing
the vector with the foreign DNA so that the 5'-phosphates are not removed from the insert
DNA. The efficiencies of removing the 5'-phosphate from overhangs (i.e., extensions), recessed
ends, or blunt ends are different and the reaction conditions need to be adjusted accordingly.
The completeness of the phophatase step is easily checked by carrying out a ligation and
transformation without insert DNA and comparing the number of colonies formed to the
number of colonies obtained with an equal amount of plasmid which was not treated with
phophatase.

Ligation of DNA Fragments into Plasmid Vectors
TERMINI
CLONING
REQUIREMENTS COMMENTS
Identical
Overhangs
Phosphatase treatment of
linear plasmid improves
efficiency.
Restriction sites at junctions preserved.
Both orientations of insert DNA possible.
Tandem copies of insert possible.
Blunt-end
High concentrations of DNA
and ligase needed.
Phosphatase treatment.
Restriction sites at junctions often
eliminated. Tandem copies of insert DNA
possible. Both orientations possible.
Different
Overhangs
Purification of double-cut
plasmid increases
efficiency.
Restriction sites at junctions preserved.
Background of non-recombinants is low.
One possible orientation of insert. Tandem
copies unlikely.

181
The vector and foreign DNA can also be prepared with restriction enzymes that produce
blunt ends. The advantage of blunt ends is that all blunt ends are compatible. However, the
restriction sites will be lost or changed when blunt ends from different restriction enzymes are
ligated together. In addition, the ligation efficiency is much lower because there are no cohesive
termini to temporarily hold the two DNA fragments together. Therefore, higher DNA
concentrations (both vector and foreign) and DNA ligase concentrations are needed for blunt-
end ligations.

Plasmids with different protruding termini can also be prepared. Advantages of different
termini are a low background of non-recombinants, tandem copies of the insert DNA are less
likely, and the insert DNA can only be in one orientation.

Transformation and Screening.

Following ligation, the rDNA is
introduced into the host. Bacteria, and in
particular E. coli, are typically used as host
cells. The two most common methods used
to transform bacteria with plasmids are
'heat-shock' and 'electroporation'. In both
cases the bacteria must be made 'competent'
(i.e., ready to receive the DNA). In the heat
shock method the bacteria are pretreated with
CaCl
2
. The competent cells are incubated
with the recombined plasmid to allow DNA
to bind to the bacteria surface and then
briefly incubated at 40-42
o
. The heating
increases the fluidity of the membrane and
Transformation Efficiency
5000 bp/plasmid
X 660 daltons/bp
= 3x10
6
g/mole plasmid
X 10
6
g/g
= 3x10
12
g/mole plasmid
1/X 3x10
-13
moles/g DNA
X 6x10
23
plasmids/mole
= 2x10
11
plasmids/g DNA
Heat Shock
(10
5
-10
9
colonies/g DNA)
Electroporation
(10
9
-10
10
colonies/g DNA)
182
supposedly opens up temporary 'pores' in the bacterial membrane that allows the DNA to enter.

The efficiency of heat shock is typically 10
5
-10
9
colonies per g DNA, which is
relatively low. For example, a 1 g of a 5 kb plasmid will contain approximately 2 x 10
11

copies. This means that even at an efficiency of 10
9
colonies per g of DNA only 0.5% of the
plasmids were taken up by bacteria during transformation. Electroporation exhibits efficiencies
of 10
9
-10
10
colonies per g DNA and can be used for the transformation of prokaryotic cells. In
electroporation, the competent cells are exposed to pulses of high voltage which will produce
transient pores in membranes and permit DNA uptake. With both electroporation and heat
shock the efficiency of transformation decreases with increasing plasmid size.

After transformation with the recombinant
plasmids the bacteria are plated on antibiotic containing
media (Figure). The antibiotic will kill the bacteria
which do not contain plasmid. Therefore, all of the
colonies will represent bacteria with plasmid. However,
the bacteria will have different plasmids and it will be
necessary to identify the clone of interest by screening the rDNA library. A common method for
screening rDNA libraries is to use a 'DNA probe' which hybridizes to the recombinant clone of
interest. Sources of DNA probes include previously cloned genes or synthetic oligonucleotides
that exhibit sequence homology to the gene of interest (Box). For example, a previously cloned
gene from another species that is homologous to the gene of interest can be used as a DNA
probe. It is also possible to make a synthetic oligonucleotide probe from highly conserved
regions of known genes. Screening expression libraries with antibodies raised against the
protein of interest is also possible (see section on expression).

In situations where a homologous gene has not been cloned or not much is known about
the gene of interest it may be possible to develop a probe based upon a partial protein sequence
(referred to as reverse genetics). A mixture of synthetic oligonucleotides, based upon all the
possible codons for the polypeptide sequence, is synthesized and used to screen the recombinant
DNA library. If possible, the amino acid sequence should be chosen to minimize the
redundancy of the genetic code (see Table). For example, methionines and tryptophans, since
they only have one possible codon, and amino acids with only two possible codons are
especially useful. Conversely, amino acids with six possible codons should be avoided.

Codons for Hypothetical Heptapeptide
Met Trp Glu Leu Ile Ala Gly
ATG (100) TGG (100) GAA (87) CTT (11) ATT (40) GCT (42) GGT (47)
GAG (13) CTA (7) ATA (53) GCA (43) GGA (46)
CTG (2) ATC (7) GCG (4) GGG (3)
CTC (2) GCC (11) GGC (3)
TTA (64)
TTG (13)

Corresponding degenerate oligonucleotide for the heptapeptide:

Sources of Probes
previously cloned genes
synthetic oligonucleotides
mixed oligonucleotides
antibodies
183
A-T-G-T-G-G-G-A-R-Y-T-N-A-T-H-G-C-N-G-G-N
1x1x1x1x1x1x1x1x2x2x1x4x1x1x3x1x1x4x1x1x4 = 768

It may be possible to decrease the number of degenerate oligonucleotides by adjusting
for the codon usage in the species of interest. Many organisms exhibit a codon-usage bias in
that not all codons are used at equal frequencies (see http://www.kazusa.or.jp/codon/ for codon
usage tables). For example, the numbers in parentheses are percentages for each the codons in
Plasmodium falciparum. Since Plasmodium species have an AT-rich genome (approximately
80%) the third codon position is very biased towards A and T and it is possible to make less
degenerate probe which still may hybridize to the target sequence.

A 'colony lift' is used to detect
bacteria containing plasmids of interest
(Figure). In this procedure, a DNA probe
is used to screen the rDNA library
utilizing blotting techniques as discussed
previously. The transformed bacteria are
grown on agar plates in the presence of
antibiotic. When bacterial colonies
become visible a sterile nylon membrane
is laid on top of the plate. Orientation
marks are made on both the membrane
and plate so that the pertinent colonies
can be identified later. The membrane is
then carefully remove and laid onto
another nutrient agar plate with the
colony side up. In other words, this membrane is a replica of the original master plate. Both the
original plate and the membrane are returned to the incubator until the colonies are the desired
size. The bacteria are lysed and the plasmid DNA is released directly onto the membrane by
laying the membrane (colony side up) on filter paper saturated with buffers containing SDS.
The released DNA will bind to the membrane and it is then denature by moving the membrane
to a filter paper saturated with NaOH. Following neutralization, the DNA is fixed to the
membrane by baking or UV cross-linking, and then incubated with a radioactive DNA probe as
described for Southern blotting. The probe will hybridize to plasmids containing homologous
insert sequences and result in a spot on the autoradiograph corresponding to the position of the
colony containing the recombinant plasmid. The positive colonies are identified on the original
plate by aligning the autoradiograph with the master plate. The bacteria can be streaked on an
agar plate and retested for the plasmid of interest to insure that the original bacteria colony was
derived from a single bacteria. Bacteria from a positive colony are expanded in liquid culture
media containing antibiotic and the recombined plasmids (i.e., cloned DNA) isolated for further
characterized.


184
BACTERIOPHAGE

Viruses can also used as cloning vectors. The ability of a virus to replicate within a host
cell is exploited as a means to amplify DNA of interest. Bacteriophages are viruses that infect
bacteria and are used in various recombinant DNA methods. Phage has proven to be an
exceptionally powerful vector for molecular cloning.

Phage Life Cycle. 1) Phage attaches to bacteria. 2) DNA is injected and circularizes
vis-a-vis cos ends. 3) DNA is transcribed and replicated*. 4) Phage assembly. 5) Cell
lysis (plaque formation). *Alternatively can intergrate into host genome and undergo a
lysogeny.

Phage infects E. coli by injecting its DNA into the bacteria (Figure). The phage DNA
is initially linear but forms a circle as a result of cohesive, or cos, ends. Phage proteins are trans-
cribed from the phage DNA and the genome is replicated. DNA replication proceeds by a
'rolling circle' mechanism that generates long linear molecules composed of consecutive
genomes. Infectious particles are then assembled from the phage proteins and DNA. The
proteins making up the phage heads self-assemble into preheads. Other phage proteins bind the
linear phage DNA near the cos site, wind the DNA into the prehead, and cleaves the DNA at the
cos sites. A fully assembled tail structure is then attached to the prehead resulting in a mature
and infectious phage particle. The phage then induces host cell lysis releasing infectious phage
185
particles which are able to infect more E.
coli.

The infection and multiplication of
phage is known as the lytic cycle.
Alternatively, can also undergo a lyso-
genic cycle in which incorporates into the
host genome at a specific site and is
replicated with the host. At a later time, the
phage genome can excise from host
genome and undergo lytic cycle producing
infectious phage particles.

Lambda as a Cloning Vector.

Two features of make it a
powerful cloning vector: 1) foreign DNA
can be inserted into the genome, and 2)
infectious phage particles can be assembled in vitro. Phage has been extensively modified to
make it a more efficient cloning vector and many different vectors have been developed. Each
particular vector has an optimal host and sometimes different applications require different
host bacteria. It is important to choose the right host strain for each particular phage and the
application being carried out.

The two basic types of vectors are insertion
and replacement vectors (Figure). The genome is
approximately 50 kb. Phage assembly can occur if the
genome is 78-105% of the normal size. In other words
infectious phage particles can be assembled if the
genome is 40-52 kb. Insertion vectors have a small non-
essential central region of the genome removed. This
removed portion of the genome allows for the insertion
of up to 10 kb of foreign DNA (depending upon the
particular vector). In replacement vectors a 13 kb
fragment, called the stuffer fragment, is excised with
restriction enzymes and discarded. Only genes necessary
for the lysogenic cycle are on the stuffer fragment and
the lytic cycle is not affected. The stuffer fragment is
replaced with a foreign DNA insert of 11-21 kb. The
insertion vectors are easier to work with but accommodate less insert DNA than replacement
vectors.

The overall strategy of cloning into lambda is similar to that of cloning into other
vectors (Box). The DNA is isolated and digested with the appropriate restriction enzyme(s).
In the case of replacement vectors the left and right arms will be purified away from the stuffer
fragment. The vector can also be dephosphorylated to minimize vector self-ligation (as
186
discussed for plasmids) when identical overhangs. Generally
vectors which have already been digested with restriction
enzymes and dephosphorylated (i.e., ready for cloning) are
obtained commercially.

The vector and foreign DNA (digested with the
appropriate restriction enzyme) are mixed and ligated. Generally,
equal molar ratios of vector and foreign DNA work best. Ligation
results in the foreign DNA being combined with the vector. In
addition, the cos ends of phage are ligated together forming very
long DNA molecules of sequential recombinant phage. The
recombined concatenated phage DNA is then mixed with an
extract containing all of the proteins needed for phage assembly. This in vitro packaging extract
is prepared from E. coli which have been infected with a defective and usually obtained from
a commercial source. One of the mutations in the defective lambda reduces the packaging of the
endogenous phage DNA, but does not affect the synthesis of the phage coat proteins. Therefore,
the recombined phage DNA is preferentially packaged into phage particles when mixed with
extracts prepared from E. coli infected with this lambda mutant. Nucleases on the phage head
recognize and cleave the DNA at the cos sites resulting in the in vitro assembly of an infectious
phage particle. The phage particles will contain different insert DNA fragments and this mixture
of recombinant phage is called a library.

The next step is to titer the recombinant library and determine the plaque forming units
(pfu) per ml. Plaques are clear zones on a bacteria lawn which are the result of localized
bacteria lysis. The bacteria lawn is formed by mixing the phage with E. coli in a low
concentration of agar or agarose. This mixture is then poured onto an agar nutrient plate and
allowed to solidify. As the bacteria grow the plate will become opaque. Infected bacteria will
result in a clear plaque due to the destruction of neighboring bacteria following many cycles of
lysis and reinfection.

It is sometimes desirable to amplify the library. Amplifying will increase the total
volume of the library allowing for multiple screenings with different probes. Amplification is
accomplished by subjecting a portion of the library to several rounds of replication before
screening the library. Libraries can be amplified by either a plate lysate method or a liquid
culture method. Both methods consist of infecting the host bacteria with the recombinant phage
and allowing the phage to replicate until the bacteria are completely lysed. In both methods the
ratio of phage to bacteria, called the multiplicity of infection (m.o.i.), has a large impact on the
final yield of phage (i.e., titer). In the plate lysate method a bacterial lawn is prepared and buffer
is then added to the plate following the complete lysis of the host bacteria to recover the phage
particles. One drawback of amplification is that the relative frequencies of the different
recombinants may change since slowly replicating recombinants will be under represented in
the amplified library. libraries are stable for years if stored at 4
o
C in the presence of
chloroform.

Cloning into
1. Prepare vector DNA.
2. Prepare foreign
DNA.
3. Ligate vector and
foreign DNA.
4. In vitro packaging.
5. Titer and amplify(?)
6. Infect host E. coli.
7. Screen plaques.
8. Plaque purification.
9. Subclone fragment
i t l id
187
Screening libraries

The library is plated out on the appropriate strain of E. coli and screened. Recombinant
clones of interest are identified with DNA probes in a plaque lift. When plaques appear, the
plates are overlaid with a nylon membrane. The phage in the plaques stick to the membranes. In
general, plaque lifts are similar to colony lifts (Figure) except that it is not necessary to continue
growth after transfer to the membranes. Phage particles are lysed directly on the membrane and
the recombinant phage DNA is denatured and fixed to the membrane. The membranes are then
hybridized with a specific DNA probe. The
recombinant phage with the insert DNA of
interest is then identified by aligning the
autoradiograph with the original plate.

The recombinant phage of interest are
recovered by punching out the plaque of
interest from the agar plate with a Pastuer
pipet. The phage are eluted from the agar and
can be expanded by reinfecting E. coli. A
'plaque purification' can be carried out by
replating and rescreening the recombinant
phage. It is recommended to repeat this plaque
purification until all of the plaques are
positive for the insert DNA of interest. The cloned phage are then expanded and the DNA
isolated.

The high efficiency of infection and the ease at which relatively large numbers of
plaques can be screened make a good vector for cloning genes. However, isolating DNA from
is somewhat cumbersome compared to isolating plasmid DNA. Therefore, the recombinant
DNA insert is usually subcloned into a plasmid for further characterization. This is easily
accomplished by digesting the phage DNA with the appropriate restriction enzyme(s) and
ligating with plasmid previously digested with the same enzyme(s). It is usually not necessary
to purify the insert fragment since the phage arms are not efficiently incorporated into the
plasmid or transfected into the host bacteria.


188
FILAMENTOUS BACTERIOPHAGE

Single-stranded DNA filamentous phage
exhibit features that can be exploited for molecular
biology research. For example, M13 has been widely
used for the generation of ssDNA for DNA
sequencing and site-directed mutagenesis. More recently, though, sequencing of dsDNA has
replaced ssDNA and many site-directed mutagenesis applications now utilize PCR. Filamentous
phage are also used for 'display' of ligands or epitopes (known as phage display).



Life cycle. Filamentous phage attach to the F-pili of E. coli and inject single-stranded
circular DNA into the bacteria. The ssDNA is converted to a dsDNA form known as replicative
form (RF). The replicative form behaves as a plasmid within the bacteria and undergoes
multiple rounds of replication. It is this replicative form that is isolated and manipulated as a
cloning vector. Foreign DNA can be incorporated into cloning sites and bacteria transformed by
standard procedures. Structural genes are transcribed from the replicative form and the ssDNA
genome is produced from a special replication origin (usually called the f1 origin). The product
of gene II forms a nick in the initiation site of the
f1 replication origin. Host DNA polymerases
replicate the phage DNA starting at the nick and
displace one strand will progressing along the
template. When the polymerase completes the
circle, the gene II product cleaves the displaced
strand at the termination site within the replication
origin and the ssDNA is circularized.

Cloning in M13
isolate RF form and manipulate
as plasmid
transfect E. coli
differential centrifugation
bacteria pellet = RF form
phage supernatant =
ssDNA
Applications of ssDNA phage
DNA sequencing
site-direct mutagenesis
phage display
in vivo excision
189
The ssDNA is packaged at the membrane of the bacteria and the mature phage is
extruded without lysing the cell. Phage particles containing ssDNA are isolated from the culture
supernatant medium by differential centrifugation and the ssDNA is extracted from the phage
particles. The phage can also be plated on a bacterial lawn to produce plaques. However, these
plaques are 'fuzzy' since they are not formed as a result of cell lysis, but formed as a result of
diminished host cell growth.

Phagemids. Many plasmids have been modified to include a f1 origin of replication.
These 'phagemids' allow for the generation of ssDNA without some of the difficulties of
working with M13. In particular, incorporation of foreign DNA into M13 will often severely
affect phage replication. Foreign DNA is incorporated into these phagemids and propagated as
conventional plasmids. Single-stranded DNA is rescued by super-infecting the transformed
cells with a helper phage (eg., R408, M13KO7). The helper phage supplies enzymes (i.e., gene
II) that are necessary for synthesis of circular ssDNA molecules and the genes for the phage
coat proteins. The helper phage, however, has a defect in its own f1 origin so that the f1 origin
of the recombinant phagemid is the preferred substrate. The ssDNA is then packaged into the
phage particles and secreted into the culture media with some contamination by the helper
phage due to its slow replication. The exact yield of the ssDNA is dependent on the nature of
the insert DNA.

ZAP is another example of a
vector exploiting the f1origin. In this
case, a phagemid (pBluescript) is
contained within the DNA such that
initiation and termination regions of
the f1 origin are at opposite ends
(Figure). Co-infection with a helper phage leads to the displacement of a ssDNA molecule from
the initiation site followed by cleavage and circularization at the termination site. The ssDNA is
then packed into phage particles and extruded from the bacteria. The plasmid containing the
foreign DNA is obtained by infecting F' strains of E. coli with these phage particles isolated
from the culture supernatant. This process, called 'in vivo excision', eliminates the subcloning
step.

Phage display. Filamentous phage have also been used for expression vectors. Peptide
or protein sequences are expressed as fusion proteins with a coat protein of the bacteriophage,
resulting in the display of the fused protein on the surface of the phage. Fusions to the N-termini
of gp3 and gp8 have minimal effects of phage infectivity. Two common applications for this
phage display are peptide libraries and single-chain antibodies. Screening in both applications is
carried out by 'biopanning'.

Peptide libraries can be used for epitope mapping, analysis of protein-protein interac-
tions, and the isolation of inhibitors or other ligands. The libraries are prepared by cloning
oligonucleotide sequences, corresponding to random peptide sequences, into the N-terminal
portion of either gp3 or gp8. The gp8 site can only accommodate peptides of approximately 6
residues.

190
Single chain antibodies, or scFv, are produced by combining the heavy chain variable
region with the light chain variable region via a linker. This artificial antibody protein can fold
such that the light and heavy chain domains come together to form a functional binding site.
Combinatorial antibody libraries are generated from naive B-cells and cloned into the gp3 site.

In both cases the desired recombinant phage are isolated by a biopanning process. The
target protein (eg., antibody for peptide libraries or antigen for scFv libraries) are absorbed to a
solid support. This target is incubated with the phage display library and non-bound phage are
wash away. The bound phage are eluted and amplified, thereby enriching for the phage which
express peptides or proteins that bind the target. This enrichment is typically 100-1000 fold,
thus 3-4 rounds of biopanning are typically carried out. (In general, the maximum complexity of
the library ranges from 10
8
-10
9
.) The individual phage in the final biopanning are then analyzed
for the peptide sequences (eg., epitope) recognized by target protein, or the scFv are isolated
and used as a reagent for subsequent analyses.

COSMIDS AND YACS

One problem with cloning vectors is their limited size as compared to the genome size
of many organisms. For example, the human genome contains more than 10
9
bp. Typical
vectors will accommodate only 10-20 kb of DNA. Similarly plasmids, although in theory can
accommodate an unlimited amount of DNA are rarely functional above 10 kb. Cosmids and
YACS are two additional cloning vectors that increase the size of insert DNA that can be
incorporated.

Cosmids are plasmids that contain the cos ends from , an E. coli origin of replication
and a selectable marker (eg., ampicillin resistance gene). Foreign DNA with a size range of 35-
45 kb is ligated with the cosmid resulting in long concatemers of DNA. The DNA is then
packaged into heads and used to infect E. coli. Even in the absence of genes the DNA will
be packaged into phage heads if the cos sites are appropriately spaced. Once inside of the
bacteria the injected DNA will form circles and replicate as a large plasmid under antibiotic
pressure.




191
Yeast artificial chromosomes (YACS) are another means to clone large fragments of
DNA. YACS are linear pieces of DNA that replicate as chromosomes within yeast cells. The
vector contains elements that allow for its replication in both
bacteria and yeast (Box). The centromere, autonomously
replicating sequences (ARS), and telomere sequences are parti-
cularly important for its function as a linear chromosome. The
bacterial elements allow for the production of the plasmid in E.
coli. Foreign DNA (100 kb-2 mb) is then incorporated so that
it is a linear piece of DNA and is then introduced into yeast
cells.


QUALITY CONTROL

The success of molecular cloning often depends upon the ability to screen large
numbers of potential recombinants. Generally the screening step is the most labor intensive of
the entire process. Therefore, it is often desirable to evaluate the quality of a DNA library before
screening. For example, in the case of cDNA libraries the average insert size is determined by
selecting a few plaques or colonies at random and determining the size of the insert DNA. This
information gives some indication of the amount of full length cDNA present in the library.
Another measurement of library quality is the percentage of non-recombinants. Many vectors
have been modified so that recombinants and non-recombinants can be easily distinguished (see
below).

In addition, the ability to distinguished recombinants from non-recombinants eliminates
the need to screen in many subcloning procedures. For example, cloning PCR fragments or
other highly enriched DNA fragments often does not require screening if recombinants are
easily identified.

Distinguishing recombinant from non-recombinant clones. One of the first methods
for distinguishing recombinants was through the loss of antibiotic resistance in pBR322.
pBR322 contains both an ampicillin resistance gene and a tetracycline resistance gene. The
foreign DNA is cloned into either the ampicillin resistance gene or the tetracycline resis-
tance gene. Replicate plates are made on either ampicillin or tetracycline. Clones that are
resistance to both antibiotics are non-recombinants, whereas recombinants will only be
resistant to one of the antibiotics.

One of the simplest examples of distinguishing non-recombinants is illustrated by
gt10. Non-recombinant gt10 in the proper hosts (eg., HflA) undergo lysogeny and do not
produce plaques. Therefore, all plaques are recombinant and non-recombinants are excluded
from the screening process.

The most common method used to distinguish non-recombinants from recombinants is
through the interruption of the -galactosidase gene. Cloning of foreign DNA into this gene will
result in the loss of its activity. The bacteria or phage are grown on nutrients plates which
contain a -galactosidase substrate which turns blue in the presence of enzyme. Colonies or
YAC Vector
E. coli origin
antibiotic resistance
yeast centromere
and ARS
ciliate telomere
yeast selectable
marker
192
plaques from non-recombinants will turn blue in the presence of 5-bromo-4-chloro-3-indoly--
D-galactoside (X-gal), whereas recombinants will remain colorless. This process is often
referred to as blue/white color selection. Some vectors (eg., gt11) utilize the entire -
galactosidase gene whereas others (eg., pUC series) only utilize a portion of the -galactosidase
gene (lacZ in Figure). The latter is called -complementation (Box).

Identifying Recombinants

Blue/White Color Screening
(-complementation)
gt10
non-recombinants undergo
lysogeny (=no plaques)
pBR322
loss of antibiotic resistance in
recombinant
gt11
interruption of -galactosidase
gene in recombinants
pUC__
loss of -complementation in
recombinants (Box)


Host expresses C-terminal portion of
-galactosidase gene on F-factor.
Vector expresses N-terminal portion of
gene interrupted by MCS (Figure).
If both are present, then -gal is active
and substrate (X-gal) is cleaved to
produce blue color.
If vector contains insert, then -gal is
inactive and colonies (or plaques)
remain colorless.




193
CHAPTER 22--EXPRESSION OF RECOMBINANT PROTEINS

One potential of recombinant DNA is the production of proteins. Many systems have
been designed for the expression of recombinant proteins. Although E. coli are easy to grow in
large quantities there are some problems associated with the expression of eukaryotic proteins
in prokaryotes. Therefore, expression systems in eukaryotic systems have also been devised. In
addition, it is also possible to use expression vectors for the cloning of genes.

General considerations in the over production of
recombinant proteins (Box) include such features as
promoters and the nature of the host for expression. Ideally
the recombinant protein should be cloned in conjunction with
a strong promoter. Promoters are elements found on the 5'-
end of genes and control their expression. A strong promoter
results in a high level of gene expression. In addition to being
strong, promoters should be regulatable so that transcription
can be turned off and on. Recombinant proteins are
sometimes toxic for the host cell and this toxicity can be
minimized by controlling the expression. The use of high copy number plasmids also increases
the production of recombinant proteins by increasing the gene dosage.

Other features are included in many expression vectors. For example, it is often
convenient to have recombinant proteins that are exported from the host cell. This can be
accomplished by engineering localization signals within the recombinant protein so that they
are directed to a particular cellular compartment. In addition, the recombinant protein can also
be expressed as a fusion protein with another protein. This will sometimes stabilize the
recombinant protein and/or assist in the purification and characterization of the recombinant
protein. Proteolysis of recombinant proteins is often a major problem that is partially alleviated
by cloning into protease deficient hosts.

EXPRESSION IN E. COLI.

The general procedure for expression
of cloned genes in E. coli involves the
insertion of the coding region of interest into
a vector, usually a plasmid, so the region is
efficiently transcribed and translated. Since
eukaryotic genes do not contain the proper
signals for transcription initiation, ribosome
recognition, translation initiation, and translation termination, these signals need to be supplied
by the vector.

Common promoters used in bacterial expression vectors are: P
L
, lac, tac and T7
(Table). Of these, the lac and tac promoters are the most widely used. The control elements that
regulate expression from these promoters are supplied by either the host or the vector. The P
L

promoter is controlled by a mutant cI repressor protein that is non-functional at 42
o
. At
Over Production of
Recombinant Proteins
strong promoter
regulatable promoter
gene dosage
localization signals
protease defective
hosts
fusion proteins
Common Bacterial Promoters
Promoter Repressor Induction
P
L
cI857 (ts) 32
o
42
o

lac lacI
q
IPTG
tac lacI
q
IPTG
T7 - T7 gene 1
194
temperatures less that 42
o
the repressor protein (cI) binds to the promoter and prevents
expression. To induce expression the temperature is raised to that the repressor becomes non-
functional and expression is now permitted. One problem with this system is that heat-shock
proteins may also be induced. The T7 promoter is from the T7 phage and is only transcribed by
the T7 RNA polymerase (T7 gene 1). The host cell must also contain the T7 gene 1 in order to
express from this promoter. To regulate expression from this promoter it is necessary to be able
to regulate the expression of the T7 gene 1.


Promoters and Repressors

The lac and tac promoters are controlled by the lac repressor (Figure). The lac repressor
binds to the lac promoter and prevents RNA polymerase from transcribing the gene. Host
strains with lacI
q
gene express the lac repressor at 10-fold higher concentrations than the normal
lacI gene. IPTG (isopropyl-1-thio--D-galactopyanoside, an analog of lactose) binds to the
repressor and prevents its interaction with the lac promoter and allowing RNA polymerase to
transcribe the regulated gene. The tac promoter is a fusion of trp and lac promoters and is also
regulated by IPTG.

The production of recombinant fusion
proteins often stabilizes the expression of foreign
proteins in E. coli. Several plasmids that express
recombinant proteins as fusion proteins have been
developed (Table). In addition to stabilizing the
recombinant protein, the fusion partner is often
exploited for affinity purification or for the analysis of
the recombinant protein (see Appendix). However, for
many applications the fusion partner may interfere
with the activity of the recombinant protein. Some
expression vectors include a protease site between the
fusion partner and the multiple cloning site (MCS). The engineered protease site allows for the
removal of the fusion partner after its purification.

Other expression vectors will only include a small fusion partner, such as epitope
tagging and His
6
vectors. In both of these examples, only a few additional amino acids are
added at either end of the recombinant protein. In general, these few residues will not interfere
with the normal activity of the recombinant protein. Affinity columns prepared from mAbs
Fusion Protein Affinity Matrix
Glutathione-S-
Transferase
glutathione-
agarose
Thioredoxin
phenylarsine
oxide-agarose
Maltose Binding
Protein
amylose-
agarose
Six Histidine
Residues
Ni-agarose
195
against the epitope can be used for the purification
of the fusion protein in the case of epitope tagging.
Similarly the six consecutive histidine residues
added by His
6
vectors bind tightly to metals and are
easily purified on metal-chelating columns.

Recombinant proteins expressed at high
levels will sometimes form insoluble aggregates
known a inclusion bodies. In some applications it
is possible to take advantage of this phenomenon.
For example, the inclusion bodies can be isolated by differential centrifugation and solubilized
under denaturing conditions (eg., urea). It is sometimes possible to renature the protein and
regain activity. In addition, fusions with E. coli thioredoxin can circumvent inclusion body
formation.

PREPARATION OF EXPRESSION VECTORS.

The 'insert' DNA must be subcloned into the expression vector so that it is in frame with
the fusion protein (or a start ATG). Most expression vectors will have a MCS with several
different restriction sites. In addition, many expression vectors are designed so that variants with
all three reading frames are available. Therefore, it is generally simple to choose restriction
enzymes that will result in a continuous open reading frame (ORF) between the fusion protein
and the foreign protein. It is also possible to shift the reading frame by cutting with restriction
enzymes, filling in with Klenow, and religating (see Appendix). Digesting with an enzyme
producing a 4-base overhang will result in a -1 frameshift and digesting with an enzyme
producing a 2-base overhang will produce a +1 shift. In addition, stop codons can be produced
by this method with certain restriction enzymes (eg., HindIII, SpeI).

SCREENING LIBRARIES WITH ANTIBODIES.

It is also possible to utilize the expression vectors as a means to screen for the gene of
interest. Recombinant DNA clones expressing proteins of interest can be detected with
antibodies, ligands or other protein activity. gt11 is a common vector for cloning genes using
antibody probes. Foreign DNA is cloned
into a lac-promoter controlled -
galactosidase gene resulting in the
expression of fusion proteins. Recombi-
nant phage expressing the protein of
interest are detected with antibodies (Box
).

A recombinant gt11 library is plated on a lawn of E. coli Y1090 and incubated at 42
o

until plaques are visible. An IPTG saturated nitrocellulose membrane is then laid on the plate
and the incubation is continued for 1-3 hr at 37
o
. This will induce the expression of -
galactosidase fusion proteins within infected bacteria. Upon lysis the recombinant proteins are
released with other bacterial proteins into the plaque and adsorbed onto the membrane. The
Screening gt11 expression libraries
1. Plate phage on Y1090 lawn.
2. Grow at 42
o
until plaques appear.
3. Overlay with IPTG saturated membrane.
4. Continue incubation 1-3 hr at 37
o
.
5. Analyze plaque lift by Western blot.
196
membrane (i.e., plaque lift) is then analyzed by Western blotting using an antibody against the
protein of interest. This will result in a spot on membrane which corresponds to a plaque formed
by the recombinant gt11 expressing the protein of interest. Such plaques are identified by
aligning the membrane with the original plate. Rabbit antisera are often problematic due to a
strong background reactivity with E. coli proteins. It is sometimes possible to pre-absorb the
anti-sera with E. coli proteins to reduce the background.

It is also possible to use a cloned recombinant gt11 phage for the production of
recombinant protein. However, this is generally more difficult than subcloning the fragment into
a plasmid and therefore not widely used. The procedure involves the production of lysogens.
The lysogens are induced to produce high levels of phage from which the fusion protein is
expressed. The fusion protein is then isolated from the infected bacteria.

EXPRESSION IN EUKARYOTES

Although expression of recombinant
proteins in E. coli is usually fairly straight
forward, it is often desirable or necessary to
express cloned genes in eukaryotes. Eukaryotic
expression systems are often needed to insure
correct folding and disulfide-bond formation,
post-translational modifications, and processing. Yeast, such as Saccharomyces cerevisiae and
Pichia species, are useful hosts for the expression of recombinant proteins since they can be
grown and manipulated like a bacteria. Pichia expression systems often allow for high level
expression of recombinant proteins using a strong alcohol oxidase promoter which is induced
by methanol. Shuttle vectors contain both a bacterial origin of replication and the origin of
replication of interest (Box). This allows for manipulations of the plasmids to be carried out in
E. coli before transforming yeast or other eukaryotes.

The two major strategies for expression of
recombinant sequences in mammalian cells are 1) stable or
transient expression of transfected DNA and 2) the use of viral
expression vectors. Mammalian and other eukaryotic cells are
able to take up DNA. The introduction of DNA to cells is
often cell type dependent and problematic, though. Several
different methods for introducing DNA to cells are available
(box). One of the early methods for transfecting eukaryotic cells was to incubate a calcium
phosphate precipitate of the DNA with the target cells. This method is relatively inefficient.
Using DEAE-dextran instead of calcium phosphate is a little more efficient. Electroporation is
the application of a short pulse of high voltage. This method is effective for many cell types
once the optimal conditions are determined. DNA can also be incoporated into liposomes
prepared from cationic detergents. These liposomes (i.e., micelles) will fuse with the plasma
membrane and deliver the DNA. Similarly, protoplasts can be prepared by digesting the
bacterial cell wall and fusing these protoplasts with the target cells. Finally, DNA can be
directly introduced into cells by ballistics (i.e, gene gun) or microinjection. The gene gun shoots
small colloidal gold particles covered with DNA into cells with doing permanent membrane
Shuttle Vectors
prokaryotic replication origin
E. coli selectable marker
eukaryotic replication origin
eukaryotic promoters/enhancers
polyadenylation signals
eukaryotic selectable marker
calcium phosphate
DEAE-dextran
electroporation
liposomes
protoplast fusion
ballistics (gene gun)
microinjection
197
damage. In many cases the foreign DNA is quickly loss from the host cell and the expression is
only transient.

Several different eukaryotic viruses can also be used as cloning vectors for the
expression of recombinant proteins. Lytic viruses are good for transient expression whereas
episomal viruses are better for constitutive expression. Retroviruses can become incorporated
into the host cell genome and possibly lead to a stable transformation.

Baculoviruses are used for the production of recombinant eukaryotic proteins.
Autographica californica is a nuclear polyhedrosis virus (AcNPV) that infects insects. Sf9 cells,
derived from Spodoptera frugiperda, are readily grown in vitro and can be infected with
baculovirus. Extremely high levels of recombinant proteins can be expressed and many
eukaryotic post-translational modifications are correctly made in the host insect cells. This high
level of expression is driven by a strong promoter for the polyhedrin gene. The polyhedrin
protein is expressed late in infection as the virus is killing the host cell and is needed for the
dissemination of the virus in nature. However, the polyhedrin gene is unnecessary for viral
growth in tissue culture and can be replaced with a gene of interest.

Baculovirus expression vectors consist of
the viral genome (130 kb) and a transfer
plasmid vector which contains the regions
flanking the polyhedrin gene. The target gene
is inserted into the plasmid vector and host
cells are co-transfected with the parental
baculovirus and the engineered plasmid.
Recombination between the viral genome and
the plasmid results in replacement of the
polyhedrin gene with the target gene. The
efficiency of identifying recombinants is
greatly enhance by digesting the baculovirus
DNA with a restriction enzyme which
disrupts an essential gene and using a transfer
vector which contains this essential gene
(Figure). Recombinants between the transfer
vector and the virus will have a selective
advantage and produce plaques. The
recombinant virus can be isolated from plaques and then amplified to produce large amounts
of protein.

198
APPENDIX 1. CODES AND CODONS

A
C
G
T or U (DNA or RNA)

M A or C (methyl)
R A or G (purine)
W A or T (weak)
S G or C (strong)
Y C or T (pyrimidine)
K G or T (keto)

V A or C or G (not T = V)
H A or C or T (not G = H)
D A or G or T (not C = D)
B C or G or T (not A = B)

N A or C or G or T (any nucleotide)




Second Codon Position
U C A G
UUU Phe UCU Ser UAU Tyr UGU Cys U
UUC Phe UCC Ser UAC Tyr UGC Cys C
UUA Leu UCA Ser UAA Stop UGA Stop A
U
UUG Leu UCG Ser UAG Stop UGG Trp G
CUU Leu CCU Pro CAU His CGU Arg U
CUC Leu CCC Pro CAC His CGC Arg C
CUA Leu CCA Pro CAA Gln CGA Arg A
C
CUG Leu CCG Pro CAG Gln CGG Arg G
AUU Ile ACU Thr AAU Asn AGU Ser U
AUC Ile ACC Thr AAC Asn AGC Ser C
AUA Ile ACA Thr AAA Lys AGA Arg A
A
AUG Met ACG Thr AAG Lys AGG Arg G
GUU Val GCU Ala GAU Asp GGU Gly U
GUC Val GCC Ala GAC Asp GGC Gly C
GUA Val GCA Ala GAA Glu GGA Gly A
F
i
r
s
t

C
o
d
o
n

P
o
s
i
t
i
o
n

G
GUG Val GCG Ala GAG Glu GGG Gly G
T
h
i
r
d

C
o
d
o
n

P
o
s
i
t
i
o
n


A GCN
C TGY
D GAY
E GAR
F TTY
G GGN
H CAY
I ATH
K AAR
L CTN
M ATG
N AAY
P CCN
Q CAR
R CGN or AGR
S TCN or AGY
T ACN
V GTN
W TGG
Y TAY
* TAR or TGA
199
APPENDIX 2. FUSION PROTEINS



The gene of interest is cloned into the expression
vector leading to the creation of a gene fusion
between the target gene and the fusion partner (FP).


Transformed E. coli are grown in large-scale liquid
cultures and after reaching the appropriate density
are induced to express the recombinant protein.



A bacterial lysate is passed over an affinity column
made from a ligand binding the fusion partner. The
recombinant fusion protein is retained by the column
and other E. coli proteins are removed. The fusion
protein is eluted with free ligand or by other means.



Many expression vectors have a specific protease site
(eg., factor Xa) engineered between the fusion
partner and the protein of interest. Treatment with
the specific protease may result in separation of the
two proteins.


The fusion partner can be separated from the the
protein of interest by repeating the affinity
chromatography since the fusion partner will be
retained by the column.


200
APPENDIX 3. GENERATING FRAMESHIFTS

Generation of a -1 Frameshift

| Sal I | | Cl aI | | HI I I |
GGT CGA CGG TAT CGA TAA GCT TGA
CCA GCT GCC ATA GCT ATT CGA ACT
Gl y Ar g Ar g Tyr Ar g *** Al a ***
Sal I
GG T CGA CGG TAT CGA TAA GCT TGA
CCA GCT GCC ATA GCT ATT CGA ACT
+ Klenow + dNTPs
GGT CGA T CGA CGG TAT CGA TAA GCT TGA
CCA GCT A GCT GCC ATA GCT ATT CGA ACT
+ DNA ligase
GGT CGA TCG ACG GTA TCG ATA AGC TTG A
CCA GCT AGC TGC CAT AGC TAT TCG AAC T
Gl y Ar g Ser Thr Val Ser I l e Ser Leu

Generation of a +1 Frameshift

| Sal I | | Cl aI | | HI I I |
GGT CGA CGG TAT CGA TAA GCT TGA
CCA GCT GCC ATA GCT ATT CGA ACT
Gl y Ar g Ar g Tyr Ar g *** Al a ***
Cla I
GGT CGA CGG TAT CGA TAA GCT TGA
CCA GCT GCC ATA GC T ATT CGA ACT
+ Klenow + dNTPs
GGT CGA CGG TAT CG CGA TAA GCT TGA
CCA GCT GCC ATA GC GCT ATT CGA ACT
+ DNA ligase
GGT CGA CGG TAT CGC GAT AAG CTT GA
CCA GCT GCC ATA GCG CTA TTC GAA CT
Gl y Ar g Ar g Tyr Ar g Asp Lys Leu

Generation of a Stop Codon

| HI I I |
AAT TCG ATA TCA AGC TTA TCG
TTA AGC TAT AGT TCG AAT AGC
I l e Ser I l e Ser Ser Leu Ser
Hind III
AAT TCG ATA TCA AGC TTA TCG
TTA AGC TAT AGT TCG A AT AGC
+ Klenow + dNTPs
AAT TCG ATA TCA AGC T AGC TTA TCG
TTA AGC TAT AGT TCG A TCG AAT AGC
+ DNA ligase
AAT TCG ATA TCA AGC TAG CTT ATC GAT
TTA AGC TAT AGT TCG ATC GAA TAG CTA
I l e Ser I l e Ser Ser ***
201
CHAPTER 23--DNA SEQUENCING

Simple and reliable DNA sequencing methods have allowed for a relatively rapid
accumulation of gene sequence data. Two methods for sequencing DNA are the dideoxy chain
termination method developed by Sanger and the chemical cleavage method developed by
Maxim and Gilbert. The dideoxy procedure is the less laborious and the most widespread. The
chemical cleavage procedure is primarily used to sequence synthetic oligonucleotides.

DIDEOXY CHAIN TERMINATION

The dideoxy chain termination method is based upon the
use of 2',3'-dideoxynucleotides (figure) as chain terminators. These
ddNTPs will be randomly incorporated into the growing poly-
nucleotide chain if DNA synthesis is carried out in a mixture of
dNTPs and ddNTPs. Dideoxynucleotides lack the 3'-OH and,
therefore, additional nucleotides cannot be added during template
mediated DNA synthesis. Since the incorporation of the ddNTPs is random the results of the
DNA synthesis will be a mixture of newly synthesized DNA fragments of various lengths with
a dideoxy-nucleotide at the 3'-end. This mixture is then analyzed by gel electrophoresis under
conditions which will resolve DNA fragments that differ in length by single nucleotides. The
sequence is then determined from the dideoxy-nucleotide which is located at the 3'-end of
progressively longer oligonucleotides.

Dideoxy chain termination requires ssDNA as a template. The ssDNA can either be
isolated from M13, phagemids, etc., or be prepared by denaturing dsDNA with NaOH. The
sequencing is more efficient and less template DNA is needed if ssDNA is used. However, it is
more laborious to prepare and isolate ssDNA from filamentous phage, and therefore, plasmid
DNA denatured with NaOH is the most widely used template. Following denaturation in
NaOH, the ssDNA is precipitated with ethanol and redissolved in an appropriate buffer.

Original Method. The original dideoxy sequencing method is based on radiolabeling the
product strand. The first step is to incubate the template ssDNA with a primer under conditions
which promote hybridization. Primers are oligonucleotides complementary to the DNA to be
sequenced. Common sequencing vectors have well characterized primer sequences that flank
the MCS. The annealed primer/DNA is mixed with DNA polymerase, -
35
S-dATP, dGTP,
dCTP and dTTP and subject to a short labeling reaction. The nucleotide concentrations used in
this labeling reaction are relatively low so that DNA synthesis proceeds slowly. This results in
the synthesis of a short segment (15-30 bases) of radioactive DNA.

The labeled DNA is then divided into four tubes containing higher concentrations of
dNTPs and either ddATP, ddGTP, ddCTP or ddTTP. This step is called the termination
reaction. The increase in the dNTP concentration result in an increased rate of DNA synthesis.
The ddNTPs are at relatively low concentrations and are incorporated at random in the growing
DNA molecule. When a ddNTP is incorporated into the growing DNA strand no additional
nucleotides can be added resulting in a chain termination. Since ddNTP incorporation is
random, the result will be a mixture of nucleotides terminated at all possible positions. The ratio
202
of ddNTP/dNTP is crucial for the success of DNA sequencing. If the ddNTP is too high then
the DNA synthesis will terminate prematurely and only sequence close to the primer will be
determined. If the ddNTP concentration is too low then insufficient oligonucleotide chains of
appropriate lengths will be produced.



The newly synthesized DNA is denatured by heating and subjected to gel electro-
phoresis under conditions which separate DNA molecules that differ in size by only a single
base. The standard conditions are polyacrylamide gels (6-8%) containing urea. The urea will
prevent H-bonding and minimizes the formation of dsDNA and secondary structures. The four
termination reactions are electrophoresed side-by-side and DNA strands are detected by auto-
radiography. The smaller oligonucleotides representing termination events that occurred near
the primer will be at the bottom of the gel. Starting at the bottom of the gel the DNA sequence
can be read directly from the autoradiograph by noting the lane (i.e., ddNTP termination
reaction) of each progressively larger band.

Semi-automated Sequencing. Pouring, running and analyzing sequencing gels are
laborious procedures. Semi-automated DNA sequencers based on fluorescent dyes are now the
preferred method of sequencing DNA. In addition, to reducing the amount of work involved in
DNA sequencing, automated sequencing yields significantly more bases of readable sequence.
The basis of automated sequencing is similar to
that of manual sequencing except that product
strand is labeled with fluorescent dyes during the
extension reaction and then detected during elec-
trophoresis. Therefore, no preliminary labeling
reaction is needed.

Sequencing reactions are carried out with
a thermocycler in a single tube containing the
template DNA, a primer, a heat stable DNA
polymerase (eg., Taq), all four dNTPs and all
four ddNTPs. The four ddNTPs that are
conjugated with different fluorochromes (i.e.,
203
different emission spectra). Use of the thermocycler does not result in a geometric amplification
as observed during PCR since a single primer is being used and the product strands are not used
in subsequent cycles. The template strand is reused in subsequent cycles though. As in manual
DNA sequencing the ddNTPs are randomly incorporated into the product strands during the
extension reaction and no additional nucleotides can be added to that particular DNA product.
The result is a mixture of DNA molecules complementary to the
template strand but differing in length by a single nucleotide and
containing one of the four possible dideoxy nucleotides at their 3'-ends.


The products of this DNA synthesis (i.e., sequencing reaction)
are subjected to gel electrophoresis in a special apparatus that
continuously analyzes the DNA molecules for fluorescence as they pass
by an excitation laser. As in manual sequencing the gel can resolve
DNA molecules that differ by a single nucleotide. These progressively
longer DNA molecules pass through this fluorometer one at a time and
the fluorescence intensity for each of the four possible emission spectra
will be determined. The wavelength of the emission spectra will
indicate which of the dideoxynucleotides was incorporated into the
terminal position of that DNA molecule. Therefore, the DNA sequence
can be directly determined by the instrument during electrophoresis. The instrument provides
the sequence of the DNA as well as a
graphical printout of the fluorescence
intensities for each of the four chromophores
plotted against nucleotide number. The
graphical printout can help resolve any
ambiguities in the sequence.

Compressions and stops. Occasionally it is
not possible to determine the DNA sequence because
of compressions and stops. Compressions are caused
by secondary DNA structure in the 'product' strand.
DNA strands containing secondary structures that are
stable in urea will migrate faster during electrophore-
sis than unfolded DNA molecules. This will result in
bands that are close together and exhibit irregular
spacing. Stops, or bands in all four lanes, are the
result of false termination during the sequencing
reaction. These stops are due to secondary structures in the 'template' strand. When the
polymerase encounters a secondary structure in the template it will dissociate without
incorporating a nucleotide. The secondary structures are often the result of dyad symmetry
and or G-G base pairing that occurs in GC rich DNA.

Several different ways to alleviated or minimize compressions and stops have been
described (Box). The simplest and usually most reliable method is to sequence the other strand.
The factors contributing to the secondary structure are typically not found on the
Alleviating Compressions
sequence 'other' strand
formamide in gel
nucleotide analogs
dITP (inosine)
7-deaza-dGTP
Taq polymerase, 72
o

formamide or DMSO in rxn
terminal transferase chase
204
complementary strand. One strategy to eliminate compressions is to include formamide in the
gel in an attempt to reduce the amount of base pairing. Another strategy is to lower the stability
of the secondary structures in the product strand by substituting ITP or 7-deaza-dGTP for GTP.
These nucleotides will prevent some of the secondary structure due to G-G base pairing. Stops
may be eliminated by minimizing the secondary structure in the template strand either before or
during the sequencing reactions. For example, using the Taq polymerase (heat stable) at 72
o

might eliminate the secondary structure in the template. It is also possible to minimize template
secondary structure by including chaotropic agents such as formamide or DMSO (up to 10%) in
the sequencing reactions. An alternate strategy in cases template secondary structure persists,
such as especially GC-rich DNA, the product strands can be treated with terminal transferase.
DNA strands that are the result of false stops will still have a 3'-OH. The terminal transferase
will add a large random number of nucleotides to the 3'-OH and thus eliminate these false stops
from the analysis since these fragments will now migrate near the top of the gel and not
interfere with readable sequence. Furthermore, the random number of nucleotides added to the
3'-end will diminish the effects on the analysis.

EXTENDED SEQUENCING STRATEGIES

The number of nucleotides that can be determined from a single sequencing reaction is
limited. Several strategies have been described for the sequencing of long DNA fragments
(Table). The choice of method(s) will depend largely and how much sequencing needs to be
done and the resources available.

Extended Sequencing Strategies
Method Advantages Disadvantages
Restriction Digestion
and Subcloning
preparation simple
contig assembly easy
relies on convenient
restriction sites
Primer Walking preparation simple slow and expensive
Nested Deletions contig assembly easy laborious to generate
Shotgun
quick and easily
automated
gaps and inefficient

The most straightforward and simplest method is to take advantage of restriction sites
within the target DNA fragment. The cloned DNA is cut with a restriction enzyme within the
MCS and a restriction enzyme within the insert fragment (Figure). If the restriction sites are not
compatible, the recessed ends are filled-in with Klenow and dNTPs. The plasmid is then re-
circularize by treating with DNA ligase, and thereby creating a deletion mutant that can be
sequenced using primers flanking the MCS. Restriction fragments can also be subcloned and
sequenced. One limitation of this method is the potential lack of conveniently located restriction
sites.

205

Sequence Restriction Fragments Primer Walking

Another strategy for extended sequencing, often called 'prime and run' or 'look and leap',
is based upon synthesizing primers based upon known sequence (Figure). In this method one
sequences a region of DNA and then designs an oligonucleotide primer corresponding to most
distal tract of reliable sequence. This new primer is
then used to sequence more of the insert DNA and
the procedure is continued until the complete
sequence is obtained. The same general rules for the
design of PCR primers also apply to the design of
sequencing primers and several computer programs
that predict potential primer sequences are
available. Generally primers should be 40-60% GC
composition and be at least 18 nucleotides in length.
In addition, primers should not fold into hairpin
loops or other secondary structures.

Prime and run strategies tend to be slow and
expensive in that one needs to continuously
synthesize primers and only a small stretch of DNA
is sequenced before the next primer can be
synthesized. In addition, it may not be possible to
identify optimal sequencing primers resulting in
primers that do not work well in the sequencing
reaction. Therefore, this method is best for filling in
small gaps in sequence information or confirming
questionable stretches of DNA sequence.

The generation of a series of nested
unidirectional deletion mutants overcomes some of
the limitations in prime and run strategies and does
not rely on the presence of convenient restriction
sites. The plasmid containing the insert to be
206
sequenced is cut with two different restriction enzymes within the MCS (Figure). The site
closest to insert DNA should leave either a blunt end or 5'-overhang (S). The other restriction
enzyme should leave a 3'-overhang (I). Many of the common sequencing vectors have poly-
linkers in which the outermost restriction sites leave 3'-overhangs.

After the plasmid is linearized, it is digested with Exonuclease III which removes
nucleotides in the 5' to 3' direction. However, ExoIII cannot digest dsDNA which has a 3'-
overhang. Therefore, treatment of the linearized plasmid with ExoIII will only result in an
unidirectional digestion toward the insert DNA fragment. The vector, including the primer site,
will be protected because of the 3'-overhang. The ratio of DNA to ExoIII needs to be carefully
controlled so that the digestion proceeds at a known rate. Aliquots are remove at defined
intervals after the start of the ExoIII digestion and the reaction is stopped. Each subsequent
aliquot will contain progressively shorter DNA fragments.

The samples are then treated with S1 (or mung bean) nuclease to remove the single-
stranded protruding 3'-ends. The DNA is then treated with Klenow to repair any 5'-overhangs
generated by the S1 treatment, insuring that the ends are blunt. Following ligation and
transformation, colonies are picked and analyzed for the size of the insert fragment remaining.
Ideally this method will yield a series of clones with inserts that are progressively shorter by an
appropriate length. The nested deletion clones are then sequenced from the primer of the
plasmid and the entire sequence is obtained from the overlapping sequences. It is somewhat
laborious to generate the series of unidirectional mutants, but once they are obtained, the
sequencing proceeds rather rapidly. The nested deletions may also be useful for other
applications.

Another method for carrying out extending
sequencing is the 'shotgun method'. In this method a
series of random overlapping clones is generated
and sequenced. Digestion of DNA with DNaseI in
the presence of Mn
2+
will randomly cut dsDNA. The
resulting fragments are then size-fractionate,
subcloned, and sequenced. The sequences are then
analyzed by computer programs that will align the
series of overlapping clones and generate the entire
sequence. The DNAse treatment is very sensitive to
DNA/DNAse ratios and often fails to generate
appropriate sized fragments. In addition, to insure
that there are no gaps in the sequence due to lost
fragments, more sequencing than necessary is
carried out. Because of these problems, this method
is not widely used in small sequencing projects. However, this method is amenable to
automation and is used in large genome projects.

207
MAXIM AND GILBERT

The Maxim and Gilbert method is based on the principal
that DNA can be specifically cleaved at certain residues. Initially
the Maxim-Gilbert method was more reproducible and
accessible than the dideoxy method. However, the dideoxy
procedure is now superior and chemical cleavage is primarily used to check the sequence of
synthetic oligonucleotides. In addition, the chemical cleavage method is sometimes used to
sequence short stretches of GC-rich DNA, which is predisposed to compressions and stops in
the dideoxy method.

The first step in Maxim/Gilbert sequencing is to end label the DNA (Box). The most
straight forward method is to use T4 polynucleotide kinase and -
32
P-ATP, resulting in a single
32
P at the 5' end (see section on labeling DNA probes). Double stranded DNA will be labeled at
both ends and therefore the two stands must be separated. Sometimes it is possible to separated
the two single strands by electrophoresis. Generally, though, such preparations will result in
preparations in which the complementary strands contaminate each other. The labeled DNA can
also be digested with a restriction enzyme and the desired fragment isolated which would result
in only one of the strands containing a radiolabel. It is also possible to label recessed 3'-ends by
filling in with Klenow and -
32
P-dNTPs. This strategy, through the choice of restriction
enzymes and radioactive nucleotide, may result in the
asymmetric labeling of the two strands.

The labeled ssDNA is placed in four or five
separate tubes containing different chemicals that
specifically cleave at certain residues (Table). Some
protocols will omit the A>C reaction. The reaction
conditions must be carefully controlled so that the
cleavages are random (i.e., partial). The samples are then analyzed by gel electrophoresis under
conditions which allow the resolution of fragments differing by a single nucleotide, as described
for dideoxy sequencing. Since the chemical cleavage reactions are not absolutely base specific,
it is more difficult to determine the sequence than in the dideoxy-chain termination method.


Chemical Cleavage
1. End label DNA.
2. Cleavage reactions.
3. Gel electrophoresis.
Base Chemical
G dimethyl sulfate
R (A+G) piperidine formate
Y (C+T) hydrazine
C hydrazine + NaCl
A>C NaOH at 90
o

208
CHAPTER 24--SEQUENCE ANALYSIS AND BIOINFORMATICS

The field of bioinformatics represents a
convergence of explosive growth in both biotech-
nology and information technology. Traditionally
bioinformatics has been synonomous with the
management and analysis of DNA and protein
sequences. The term now is more broadly used to
include epidemiology (especially genetic
epidemiology) and evolutionary studies. One
aspect of computational biology research is the
development of new algorithms and statistics.
The other aspect of bioinformatics is the use of
these tools for the analysis and interpretation of
sequence data.

DNA sequencing produces large amounts
of data that would be extremely tedious to analyze without computers. Several commercially
available software packages will carry out routine sequence analyses (Box). Many sequence
analysis programs are also available on the internet (see Appendix). Some of these programs
can be downloaded and used on a free standing computer. In other cases the sequence is
submitted via the website or by email and the results of the analysis are provided.

A DNA sequencing project will require the use of a computer to assist in compiling and
editing the sequence data. Overlapping clones are found and assembled into contiguous
sequences. Computers are also used to generate the complementary sequence, reverse the 5'3'
orientation, or provide amino acid translations of nucleic acid sequences. It is also possible to
calculate various physical, biochemical or structural properties of both nucleic acids and
proteins. For example, algorithms to predict the secondary structure of either proteins or RNA
are available. Specific sites or motifs within a gene, such as restriction sites, promoter elements
or other DNA signals, as well as various protein motifs (eg., glycosylation sites, signal
sequences, etc.) can be identified with the computer.

Alignment of related sequences and homology searches are common analyses
performed on both nucleic acid and protein sequences. For example, it is relatively easy to
obtain sequence data, but it is quite difficult to predict protein structure and function. Protein
structure and function can be approximated from similarities to known sequences. Such
similarities are identified by aligning DNA or protein sequences.

SEQUENCE ALIGNMENT

Alignments provide a powerful way to compare sequences for either evolutionary
relatedness or structural/functional relatedness. Sequences can be compared by either global or
local alignments. Global alignment forces complete alignment of the input sequences, whereas
local alignments will only align their most similar segments. The choice of method will depend
on whether the sequences are presumed to be related over their entire lengths or to only share
Common Sequence Analyses
compiling and editing sequences
manipulating sequences
complement, reverse, or both
translating
structural analysis
searching for coding regions
physical properties
comparisons and homology
searches
aligning sequences
sites and motifs
gene identification
209
isolated regions of homology. The algorithms to carry out global and local alignments are
similar, but the statistics associated with the output are different.


Pair-wise sequence alignments use a scoring
matrix to calculate the best alignment between two
sequences. A scoring matrix contains information on
how to score each match in an alignment and
penalties for mismatches. For example in DNA
alignments, matches are assigned a value of 0.9 and mismatches a value of -0.1. Proteins are
more complicated and the scoring matrix must also take into account similarity between
residues and their relative abundance. For example, tyrosine and tryptophan residues are
uncommon and weighted heavier that alignments between common residues such as alanine
and serine. Similarly, alignments between glutamate and aspartate are scored positively
because of their chemical similarity, whereas matches between dissimilar residues are
penalized. Furthermore, amino acid similarity can be defined as similar on either a chemical
basis or a structural basis (Table). No single scoring matrix can be universally used for
proteins. BLOSUM and PAM are two well known scoring matrices.

The gap penalty is another parameter used in sequence
alignments. Mathematically, the penalty for opening a gap of
length k is defined as:

W
k
= a + bk

where a is the gap opening penalty and b is the gap
extension penalty. Both the opening penalty and the extension
penalty can be varied and it may be useful to run alignments with different values unless it is
already known the types of matches being searched for. For example, having a large open
and a large extend penalty is good for analyzing closely related proteins. A large open
penalty and small extend penalty may be good for situations where the distance between
domains is not crucial. A small open penalty and large extend penalty (ie, many small
inserts) is useful for distantly related homologous proteins.

Aligning sequences involves submitting the two sequences, choosing the scoring
matrix and prescribing the gap penalties. The alignment giving the best score for a particular
scoring matrix and the prescribed gap penalties is returned. However, this is not necessarily
the most biological significant alignment. Quite often finding the optimal alignment
Amino Acid Similarities
Chemical Physical
A, G
D, E
F, Y
K, R
I, L, M, V
Q, N
S, T
C, S
D, L, N
E, Q
F, H, W, Y
I, T, V
K, M, R
Homologs are structures or objects that share a common evolutionary
origin. Objects with similar structure or function, but no common
ancestor are analogs. Homologs can be classified as orthologs or
paralogs. Orthologs are homologous genes from different species that
arose from a common ancestor gene. Paralogs are homologous genes
that are the result of gene duplication within an evolutionary lineage
(eg., species) and may have different or similar functions.
GCGCCTC
| | | | |
GCGGGTC
(5 x 0.9) +
(2 x -0.1) = 4.3
210
involves carrying out several alignments using different values for these parameters. In
addition, the final alignment often involves using human insight.

In contrast to pair-wise alignments, which always return the best possible mathema-
tical match, multiple alignments are a first approximation. The human eye coupled with
biological insight is much better at spotting patterns in multiple alignments that the currently
available computer algorithms. Alignments are a non-trivial computational task and it is best
not to treat them like a black box programs

SEARCHING DATABASES

Comparing a sequence against a database to discover similarities is one of the most
frequently used and powerful tools in bioinformatics. In essence, searching a database is the
same as aligning two sequences. However, to perform a pair-wise alignment of the query
sequence with every sequence in the database
would be too time consuming. Therefore
programs use heuristic strategies to speed up the
analysis are usually used. FASTA and BLAST
are the two most commonly used programs for
searching databases. Both programs use rapid
exact match procedures to first identify
sequences which are most likely to be related.
These sequences are then subjected to further
analyses by alignment programs. BLAST (basic
local alignment search tool) uses an approach
based on matching short sequence fragments and then finds the best local alignments between
the query sequence and the database sequences. Several BLAST programs are available for
different types of sequence matching (see Table).

BLAST PROGRAMS
PROGRAM QUERY DB COMMENTS
BLASTP protein protein
compares amino acid query against protein
sequences
BLASTN DNA DNA
compares nucleotide query against DNA
sequences
BLASTX DNA protein
compares 6X translations of nucleotide query
against protein sequences
TBLASTN protein DNA
compares protein query against 6X
translations of DNA sequences
TBLASTX DNA DNA
compares 6X translations of nucleotide query
against 6X translations of DNA sequences

BLAST searches of the most current version of GenBank can be performed at the
National Center for Biotechnology Information (NCBI) by cutting and pasting the query
sequence onto their web page (http://www.ncbi.nlm.nih.gov/ BLAST/). The entire database or
Databases
primary (original biological data)
vs. secondary (value added)
three 1
o
DNA databases
GenBank
EMBL
DDBJ
subdivisions
annotated
211
subdivisions of the database can be searched. It is also desirable to 'filter' low complexity
sequences such as long runs of repetitive sequence (eg., tracts of poly-alanine, etc.) and most
BLAST programs filter by default. Such regions of sequence can give spuriously high scores
against unrelated proteins. Similarly, iterative searches are prone to contamination by regions
corresponding to coiled-coils or transmembrane domains. These characteristics might match the
initial search and the program then emphasizes these inappropriate characteristics. The user can
also choose cut-off values (i.e., E-values), the number of matches to report, and the number of
alignments to show.

The entire database(s) is searched and sequences with similarity to the query sequence
are identified and ranked accordingly (see Appendix 2 for sample output) to the alignment score
derived from the pair-wise alignment. In addition to the this 'raw or bit score', the output
includes a 'statistical score' and the alignments. The statistical score, or expectation value
(E-value), provides a measure of the expected number of sequences in the database that would
achieve a given alignment score by chance. The E-values are more useful in terms of judging
the significance of a match than the alignment score. E-values for proteins and DNA should be
less than 0.1 and 0.0001, respectively, to be considered significant. Examining the alignments
will also help in inferring whether the hit is significant. For example, protein global alignments
which show 25% or greater sequence identity over a stretch of at least 80 amino acids should
exhibit the same basic structure. The identified sequences can be also retrieved and subjected to
further analyses. Database searches may assist in identifying an unknown gene or provide clues
about its function.

PROTEOMICS

The availability of the complete genomic sequence from several organisms repre-
sents a major advancement in the understanding of biology. However, the gene sequences
provide a limited amount of information about the proteins that are encoded by the genes. In
addition, phenotype will be determined by the proteins and not the genes (i.e., genotype).
Efforts are now being directed towards the characterization of the proteome, or the complete
set of proteins found in a cell, tissue or organism. Proteomics encompasses many types of
activities and analyses with a goal of having a complete understanding of protein function
and to apply this knowledge (Figure).

212
One objective of proteomics is to identify the complete complement of proteins.
Genomics provides a first step in this process since all of the genes within a genome can
potentially be identified. However, the algorithms used to predict the exon/intron structure
of a gene, and to the lesser extend the beginning and end of a gene, are not always accurate.
In addition, some genes can undergo alternate splicing and thus produce multiple proteins
from the same gene. Therefore, many of the predicted genes will need to be confirmed by
additional analysis.
The proteome is a dynamic entity in that genes are expressed at specific times and
places and environmental conditions will also influence gene expression. Furthermore, many
proteins are found in specific subcellular compartments or modified post-translationally
(eg., phosphorylation, glycosylation, acylation, proteolysis, etc.). The subcellular location of
a protein as well as the post-translational modifications will have an impact on cellular
phenotype. Protein function is also dependent on protein-protein interactions and resolving
these networks of protein-protein interactions is also central to understanding protein
function and cellular phenotype. Knowing the structure of proteins also contributes to
understanding protein function. Protein structure can be determined by X-ray
crystallography and biophysical techniques. Some elements of protein structure can also be
inferred from the protein sequence and comparing these sequences to related proteins in
which the structures have been determined. The information about a protein's structure and
function can then be used in applications such as drug development.
Gene identification and expression. Some information on proteins is already
available due to previous research and protein characterization and this information is easily
incorporated into the genome databases. Similarly information on the identity and
expression of proteins will be continuously generated as particular proteins are studied.
However, this approach is selective in that proteins and genes which are associated with
interesting phenomena will be preferentially studied and it will take substantial time to
individually characterize all of the genes with unknown function.
Expressed sequence tags (ESTs) and DNA microarrays provide a more rapid means
in regards to characterizing genes and their expression. ESTs represent a collection mRNA
sequences (i.e., cDNA sequences) maintained in DNA sequence databases which have been
cloned from a defined source and partially sequenced. The cDNA clones have been chosen
at random from a cDNA library without selection and represent the mRNA profile of the
source (i.e., organism, tissue, cell, etc.) and thus provide information about gene expression.
These ESTs can also be used to assist in the identification of expression start sites and
positions of introns, even in the cases of unknown genes, by aligning the genomic sequence
with the EST sequence. The original clones from which the EST sequences were obtained
are maintain in respositories and can be obtained and sequenced in their entirities. Micro-
array technology (see section on DNA Microarrays in Hybridization Chapter) is being used
to identify the expression of mRNA on a large scale.
Protein identification. However, in many cases the genes of specific proteins have
not been identified and the presence of mRNA does not always correlate with protein
expression. The evaluation of proteins on a large scale, or by high throughput methods, is
213
not as advanced as the generation and analysis of DNA sequence. Much of protein analsysis
still involves the analysis of one protein at a time and requires the separation, identification
and characterization of proteins resolved from complex mixtures.
Gel electrophoresis of proteins is the predominant technique used in the separation
and isolation of proteins (see Chapter on Protein Electrophoresis). One-dimensional SDS gel
electrophoresis is capable of resolving many complex mixtures of proteins into individual
polypeptide chains. Two-dimensional gels (IEF + SDS-PAGE) can be used if additional
resolution is needed. Proteins are then be transferred to a membrane and subjected to
N-terminal microsequencing (see Microsequening section in Protein Purification Overview
chapter). In cases where the N-termini are blocked it may be possible to break the protein
into smaller peptides with site-specific proteases or chemicals and determine an internal
sequence. However, substantially more protein is needed to determine these internal
sequences.
Problems with blocked N-termini and low abundance may be alleviated by a mixed
peptide approach. In this method proteins are subjected to gel electrophoresis, transferred to
a membrane and the band of interest excised as in a typical microsequencing protocol. The
isolated protein is then cleaved into peptides with CNBr (cleaves at Met) or skatole (cleaves
at Trp). On average 3-5 peptides will be generated depending on the frequency of these
these amino acids. The membrane is then subjected to 6-12 automated Edman cycles. Every
cycle will result in several amino acid residues. This mixed sequence data are then analyzed
by a computer algorithm which will sort and match the data against protein or DNA
databases to identify candidate protein(s). (Additional information on these programs is
available at http://fasta.biochem.virginia.edu/)
Mass spectrometry. An increase in the sensitivity of protein identification can be
achieved with a mass spectrometer. Mass spectrometers consist of three basic components:
an ionization source, one or more mass analyzers, and an ion detector. Various types of
instruments utilizing different types of ionization sources and mass analyzers exist.
Biological samples are ionized through the addition or loss of protons. Following ionization
the masses of the components are determined. For example, in a time-of-flight mass
analyzer the time it takes to traverse the length of the flight tube is a function of mass. A
spectrum of the components plotted according to mass is recorded.

The mass spectrometer can be used to determine accurate peptide masses or amino
acid sequences of peptides. This information is then be used to identify proteins by
searching databases. For example, a protein is digested with trypsin, or another site specific
214
protease, and the resulting peptides analyzed by mass spectrometry. This acquired MS
spectrum can then be compared with predicted MS spectrums of all the proteins in a
database. The advantage of this method is that the analysis can be carried out quickly and
can be completely automated. However, there is some ambiguity in protein identification
due to peptide mass redundancy (i.e., same amino acid composition but different sequence
will be the same mass). In addition, factors which affect the accurate determination of
peptide mass (eg., post-translational modifications) will limit the abililty to successfully
identify proteins in databases.

Tandem mass spectrometry (MS/MS) can be used to deduce the amino acid sequence
of a peptide. In this approach ions with particular masses (i.e., m/z values) are separated
from the other ions and then fragmented in a collision chamber. The collision chamber is
filled with an inert gas (usually argon or nitrogen) and collisions with the ions and the gas
atoms result in covalent bonds being broken. The masses of the resulting fragments are then
determined by a second mass analyzer. The mass spectrum of the fragments is diagnostic of
the molecular structure of the parent ion. For example, peptides tend to fragment primarily
along the peptide backbone at the amide bond which creates a ladder of ions that is indica-
tive of the amino acid sequence. The sequence is used to search databases. Multiple peptides
from the same protein can be analyzed in this fashion and thus increasing the confidence of
the protein identification. The disadvantage of MS/MS is that the process is not easy to
completely automated. Although computer programs are available, interpretation and analy-
sis of the mass spectrum requires some human guidance.
High throughput analyses. The procedures described above rely upon identification
of proteins by gel electrophoresis followed by their subsequent identification using either
one- or two-dimensional gel electrophoresis. The ability of two-dimensional gels to resolve
215
thousands of polypeptides on a single gel makes it a powerful technique in proteomics.
However, 2-D gels are somewhat limited to the identification of relatively abundant proteins
in the mid-to-low molecular weight ranges. Furthermore, membrane associated proteins and
cytoskeletal proteins are difficult to solubilize and hydrophobic proteins do not always
resolve well in the isoelectric focusing dimension. In addition, two-dimensional gel
electrophoresis is a labor intensive and time consuming process and the protein 'spots' need
to be further characterized by mass spectrometry or Edman degradation individually.
To overcome these limitations high throughput methods to analyze complex
mixtures of proteins are being developed. One approach is to combine a high throughput
microcaillary liquid chromatography with tandem mass spectrotometry. In this approach
complex mixtures of proteins are treated with site-specific endoproteases. The resulting
peptides are subjected to liquid chromatography (HPLC) without a prior gel electrophoresis
step and then directly introduced into the tandem MS. Elimination of the gel electrophoresis
step provides provides a greater amount of flexibility in the proteolytic digestion and also
increases the overall sensitivity. The first MS analysis will determine an accurate mass of
the peptides which is indicative of the amino acid composition and then the fragmentation of
those peptides can be used to predict the amino acid sequence. Databases are then searched
for potential matches.
This LC/MS/MS approach can be used to generate protein expression maps that are
analogous to EST databases. Crude proteins from a particular tissue can be fractionated into
cytosolic, membrane/organellar, cytoskeletal, etc. fractions and then analyzed for the
expressed proteins in the mixture. This type of analysis will work best in organisms in
which the genome sequence or large banks of EST sequences are available. This approach
can also be used in the identification of the components of less complex protein mixtures
such as large multi-protein complexes. Similarly protein-protein interactions can be
discerned by affinity purifying all of the proteins that interact with the target protein and
then analyzing the mixture. Methods to purify binding proteins include: co-
immunoprecipitation with antibodies against the target protein, co-precipitation with an
affinity tagged (eg., His
6
) recombinant target protein, or affinity chromatography with the
target protein.
Additional Reading on Proteomics
Yates, J.R. 2000. Mass spectrometry from genomics to proteomics. Tr. Genet. 16:5-8.
Graves, P.R. and Haystead, T.A.J.. 2002. Molecular biologist's guide to protomics.
Microbiol. Mol. Biol. Rev. 66:39-63
216
APPENDIX 1. WEBSITES

List of links to molecular biology sites

http://www.expasy.ch/alinks.html

Sequence retrievial system (capable of searching numerous databases and medline
simutaneously)

http://www.ncbi.nlm.nih.gov/Entrez/

Database Searches (submit a query sequence and search for homologies in the databases)

http://www.ncbi.nlm.nih.gov/BLAST/
http://www2.ebi.ac.uk/fasta3/

Translation (translate nucleotide sequence into protein sequence)

http://www.expasy.ch/tools/dna.html

Pairwise Alignments

http://genome.eerie.fr/fasta/align-query.html

Multiple Alignments

http://www.ibc.wustl.edu/service/clustal.html
http://dot.imgen.bcm.tmc.edu:9331/multi-align/Options/clustalw.html

Protein Analysis (numerous programs for analyzing proteins)

http://www.expasy.ch
http://www.ebi.ac.uk/Tools/index.html

2
o
Protein Structure

http://www.embl-heidelberg.de/predictprotein/predictprotein.html


217
APPENDIX 2. RESULTS OF BLAST SEARCH



Quer y= Pbpp58b ( 423 l et t er s)

Dat abase: nr ( 493, 611 sequences; 154, 780, 071 t ot al l et t er s)

Scor e E
Sequences pr oduci ng si gni f i cant al i gnment s: ( bi t s) Val ue

sp| Q08168| HRP_PLABE 58 KD PHOSPHOPROTEI N ( HEAT SHOCK- RELATED PRO. . . 334 1e- 90
gb| AAC37300. 1| ( L21710) 58 kDa phosphopr ot ei n [ Pl asmodi umber ghei ] 329 3e- 89
pi r | | T10455 heat shock r el at ed pr ot ei n - Pl asmodi umber ghei >gi | . . . 250 2e- 65
sp| P50503| HI P_RAT HSC70- I NTERACTI NG PROTEI N >gi | 4379408| emb| CAA5. . . 106 5e- 22
sp| P50502| HI P_HUMAN HSC70- I NTERACTI NG PROTEI N ( PROGESTERONE RECE. . . 87 3e- 16
gb| AAF45894. 1| ( AE003429) CG2947 gene pr oduct [ Dr osophi l a mel ano. . . 87 4e- 16
pi r | | T24865 hypot het i cal pr ot ei n T12D8. 8 - Caenor habdi t i s el egan. . . 86 5e- 16
pi r | | T04562 hypot het i cal pr ot ei n T12H17. 60 - Ar abi dopsi s t hal i an. . . 81 2e- 14
.
.
.
.
.
gb| AAC60555. 2| ( S59774) STI 1 st r ess- i nduci bl e pr ot ei n homol og [ S. . . 48 2e- 04
gb| AAD33401. 1| AF129086_1 ( AF129086) car boxy t er mi nus of Hsp70- i n. . . 48 2e- 04
pi r | | A56534 P58 pr ot ei n - bovi ne >gi | 468012| gb| AAA17795. 1| ( U046. . . 48 2e- 04
gb| AAB49720. 1| ( U89984) t r ansf or mat i on- sensi t i ve pr ot ei n homol og. . . 47 4e- 04
pi r | | T16689 hypot het i cal pr ot ei n R05F9. 10 - Caenor habdi t i s el ega. . . 47 4e- 04
gb| AAD33400. 1| AF129085_1 ( AF129085) car boxy t er mi nus of Hsp70- i n. . . 46 6e- 04
.
.
.
.
.
.
emb| CAA61595. 1| ( X89416) pr ot ei n phosphat ase 5 [ Homo sapi ens] 43 0. 007
pdb| 1A17| Tet r at r i copept i de Repeat s Of Pr ot ei n Phosphat ase 5 43 0. 007
r ef | NP_006238. 1| | pr ot ei n phosphat ase 5, cat al yt i c subuni t >gi | 1. . . 43 0. 007
pi r | | S52570 phosphopr ot ei n phosphat ase ( EC 3. 1. 3. 16) 5, cat al yt i . . . 43 0. 007
gb| AAB18614. 1| ( U12203) phosphopr ot ei n phosphat ase [ Rat t us nor ve. . . 43 0. 007
sp| P53042| PPP5_RAT SERI NE/ THREONI NE PROTEI N PHOSPHATASE 5 ( PP5) . . . 43 0. 007
gb| AAB60384. 1| ( U25174) ser i ne- t hr eoni ne phosphat ase [ Homo sapi ens] 43 0. 007
sp| Q60676| PPP5_MOUSE SERI NE/ THREONI NE PROTEI N PHOSPHATASE 5 ( PP5. . . 42 0. 009
gb| AAB70573. 1| ( AF018262) pr ot ei n phosphat ase 5; PP5 [ Mus muscul us] 42 0. 009


Examples of Alignments


>sp| P50503| HI P_RAT HSC70- I NTERACTI NG PROTEI N >gi | 4379408| emb| CAA57546. 1| ( X82021)
Hsc70- i nt er act i ng pr ot ei n [ Rat t us nor vegi cus] ( Lengt h = 368)

Scor e = 106 bi t s ( 261) , Expect = 5e- 22
I dent i t i es = 60/ 224 ( 26%) , Posi t i ves = 97/ 224 ( 42%)

Quer y: 1 MDI EKI EDLKKFVASCEENPSI LLKPELSFFKDFI ESFGGKI KKDKMGYXXXXXXXXXXX 60
MD K+ +L+ FV C ++PS+L E+ F ++++ES GGK+
Sbj ct : 1 MDPRKVSELRAFVKMCRQDPSVLHTEEMRFLREWVESMGGKVPPATHKAKSEENTKEEKR 60

( SDEEEEDEEEEEEEEEDDDPEKLE)
Quer y: 61 XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXAVECPPLAPXXXXXXXXXXXXXXCKLKEEA 120
+ P + K A
Sbj ct : 61 DKTTEDNI KTEEPSSEESDLEI DNEGVI EADTDAPQEMGDENAEI TEAMMDEANEKKGAA 120

Quer y: 121 VDLVENKKYEEALEKYNKI I SFGNPSAMI YTKRASI LLNLKRPKACI RDCTEALNLNVDS 180
+D + + + ++A++ + I A++Y KRAS+ + L++P A I RDC A+ +N DS
Sbj ct : 121 I DALNDGELQKAI DLFTDAI KLNPRLAI LYAKRASVFVKLQKPNAAI RDCDRAI EI NPDS 180
Filtering--Glu rich
Entries removed
for clarity
Entries removed
for clarity
Probability of match
appearing by chance
Alignment Score
Identifier Line:
database | accession # | name or locus
218

Quer y: 181 ANAYKI RAKAYRYLGKWEFAHADMEQGQKI DYDENLWDMQKLI Q 224
A YK R KA+R LG WE A D+ K+DYDE+ M + +Q
Sbj ct : 181 AQPYKWRGKAHRLLGHWEEAARDLALACKLDYDEDASAMLREVQ 224





>pi r | | T24865 hypot het i cal pr ot ei n T12D8. 8 - Caenor habdi t i s el egans ( Lengt h = 422)

Scor e = 86. 2 bi t s ( 210) , Expect = 5e- 16
I dent i t i es = 44/ 101 ( 43%) , Posi t i ves = 60/ 101 ( 58%) , Gaps = 2/ 101 ( 1%)

Quer y: 119 EAVDLVENKKYEEALEKYNKI I SFGNPSAMI YTKRASI LLNLKRPKACI RDCTEALNLNV 178
+A + N ++ AL + I SAM++ KRA++LL LKRP A I DC +A+++N
Sbj ct : 121 KAQEAFSNGDFDTALTHFTAAI EANPGSAMLHAKRANVLLKLKRPVAAI ADCDKAI SI NP 180

Quer y: 179 DSANAYKI RAKAYRYLGKWEFAHADMEQGQKI DYDE- - NLW217
DSA YK R +A R LGKW A D+ K+DYDE N W
Sbj ct : 181 DSAQGYKFRGRANRLLGKWVEAKTDLATACKLDYDEAANEW221


Scor e = 41. 4 bi t s ( 95) , Expect = 0. 016
I dent i t i es = 16/ 34 ( 47%) , Posi t i ves = 23/ 34 ( 67%)

Quer y: 9 LKKFVASCEENPSI LLKPELSFFKDFI ESFGGKI 42
LK+FV C+ NP++L PE FFKD++ S G +
Sbj ct : 7 LKQFVGMCQANPAVLHAPEFGFFKDYLVSLGATL 40




Middle row = matches and similar residues (+)
Gaps to maximize alignment
A second high-
scoring segment